Method, device and equipment for optimizing patrol path of unmanned aerial vehicle in smart park

Abstract

The invention relates to the technical field of flight path control, in particular to a smart park unmanned aerial vehicle patrol path optimization method, device and equipment, and the method comprises the following steps: obtaining the three-dimensional coordinates of task points and analyzing the spacing trend, screening dense path segments to extract a task set, adjusting the path structure to form a distribution set, and carrying out the optimization of the patrol path. And identifying the mark change of the disturbance path section in the high-temperature area, and judging the distribution balance optimization path structure. According to the method, the three-dimensional coordinates of the task points are obtained, path spacing changes are recognized, dense sections are marked to enhance spatial perception, set screening is achieved in combination with the gravity center and boundary relation, path division independence is improved, attitude data are extracted, the disturbance trend is recognized, and the wind disturbance area is positioned; the method comprises the steps of uniformly adjusting offset nodes, complementing position information, optimizing path stability, extracting path parameters, analyzing distribution balance, dynamically reconstructing a path structure, realizing multi-path load coordination, covering data perception, structure identification and scheduling optimization by overall processing logic, and improving flight efficiency and path adaptability.

Description

Technical Field

[0001] The present invention relates to the field of flight path control technology, and in particular to a method, device and equipment for optimizing the inspection path of a drone in a smart park. Background Art

[0002] The field of flight path control technology involves the control and regulation of non-electrical variables such as the position, speed, and heading of an aircraft in space. Its core issues include aircraft attitude stabilization, heading maintenance, trajectory tracking, and path planning. This technical field mainly establishes an aircraft dynamic model, designs control laws, and implements a path generation algorithm to achieve autonomous navigation and path control of the aircraft. This field is widely used in scenarios such as civil drones, industrial inspections, logistics and transportation, and military reconnaissance. Among them, the traditional smart park drone inspection path optimization method refers to the arrangement and optimization of the path of drone inspection tasks in the park environment to improve coverage efficiency and operational rationality. Graph theory-based path planning methods are usually used to solve the targeted technical issues. For example, after constructing a park area graph model to obtain the relationship between inspection points, a depth-first search or greedy algorithm is used to traverse and sort the path to achieve preliminary planning and optimization of the drone inspection path.

[0003] In practical applications, traditional path optimization methods suffer from over-reliance on regional graph models, a single path traversal algorithm, and low spatial parameter utilization. They often map inspection point relationships using a graph structure, ignoring the three-dimensional correlations between task points in their spatial distribution. This makes it difficult to capture the dense fluctuations within path segments, resulting in a coarse path structure and blurred work interval boundaries, which impacts path scheduling accuracy. Path allocation fails to fully consider the coupling relationship between task point clustering trends and boundary locations, which can easily lead to overlapping or duplicated flight missions. When addressing high-temperature and other disturbed areas, the lack of a path perturbation detection mechanism based on attitude-related parameters results in delayed path adjustments and flight trajectory deviations, further impacting coverage integrity in critical areas. Existing path coordination mechanisms, based on static graph models, are unable to dynamically assess the degree of coordination between the number and distribution of task points within a path. This makes task loads susceptible to deviations, leading to reduced collaborative flight efficiency. These issues are particularly prominent in scenarios where multiple aircraft are performing high-density inspections, making it difficult to meet the high demands of actual operations for path rationality, node stability, and scheduling efficiency. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a method, device and equipment for optimizing the inspection path of smart park drones.

[0005] In order to achieve the above objectives, the present invention adopts the following technical solution: a method for optimizing the inspection path of a smart park drone, comprising the following steps:

[0006] S1: Obtain the three-dimensional coordinates of the patrol task points in the closed area channel, identify the distance change trend between the task points in the path segment, extract the average spacing value, compare it with the preset spacing threshold, determine the path segment area where the change range exceeds the threshold interval, and establish the path segment density recognition result;

[0007] S2: extracting a task point set based on the density identification result of the path segment, determining the positional relationship between the set center and the task segment boundary, screening the independence of the task set, adjusting the path structure of the substation fence and high-voltage line tower area, and establishing an inspection path allocation set;

[0008] S3: Calling the high-temperature area path segment of the thermal pipeline in the plant area from the inspection task allocation path set, extracting the flight attitude correlation data of the inertial monitoring points in the path, identifying and marking the path segment with continuous changes within the monitoring period, and establishing the path disturbance impact segment identification information;

[0009] S4: Extract the affected path segment based on the path disturbance impact segment identification information, analyze the angular relationship between the propulsion direction and the wind disturbance direction, and determine whether the offset value exceeds the deviation range. If the conditions are met, uniformly adjust the node positions within the path segment and establish a list of path point position adjustments after the disturbance.

[0010] As a further solution of the present invention, the path segment density identification result includes the change amplitude of the task point spacing, the area of ​​the path segment exceeding the threshold, and the area label information; the inspection task allocation path set includes the independence screening task point set, the path connection structure, and the initial path division result; the path disturbance affected section identification information includes the abnormal change section position, the wind disturbance direction trend, and the summary information of the disturbed path segment; the post-disturbance path point position adjustment list includes the node position adjustment value, the key node position information, and the adjusted path node set.

[0011] As a further solution of the present invention, the specific steps of S1 are:

[0012] S101: Obtain the three-dimensional coordinate information of the inspection task points, construct a path segment, and sequentially extract the three-dimensional coordinate values ​​between adjacent task points. Calculate the distance between the task points based on the three-dimensional coordinate values ​​to generate the path segment distance value between the task points.

[0013] S102: Calling the path segment distance value between the task points, calculating the average spacing value between the task points in the path segment, and performing a difference judgment with the set spacing threshold, extracting the path segment number and task point number that exceed the threshold, and obtaining the path segment spacing exceeding limit number value;

[0014] S103: Based on the path segment spacing overrun number value, adjacent overrun path segments are combined, start and end task point numbers are extracted, path segment area boundaries are constructed and annotated, and a path segment density recognition result is established.

[0015] As a further solution of the present invention, the specific steps of S2 are:

[0016] S201: Obtain a trajectory node set in the path segment density recognition result, calculate the centroid position of the path segment in the set using the node coordinate values, and generate a path segment centroid position coordinate set;

[0017] S202: Based on the coordinate set of the center of gravity position of the path segment, extract the spatial distance between the center of gravity point and the boundary point of the task segment, determine whether the center of gravity point is outside the boundary distance threshold range, and obtain a set of independent task path segments;

[0018] S203: According to the path connection relationship between the independent task path segment set and the fence area and the high-voltage line tower area, the connection structure between the path segment and the regional node is detected, a multi-path segment connection matrix is ​​established, and a multi-UAV collaborative path segment set is generated.

[0019] As a further solution of the present invention, the specific steps of S3 are:

[0020] S301: Acquire inertial monitoring point data in the collaborative path segment set of the multiple UAVs, combine the flight attitude information of each UAV, integrate and analyze the data of each monitoring point in the path segment, and generate flight attitude data;

[0021] S302: Based on the flight attitude data, determine abnormal fluctuations within a monitoring period using a set threshold, identify abnormal change segments in the path segments, and generate abnormal change identifiers;

[0022] S303: Analyze the changing trends of the disturbed path segments and wind disturbance directions according to the abnormal change identifiers, mark the positions and directions of the disturbed path segments, and generate path disturbance impact segment identifier information.

[0023] As a further solution of the present invention, the specific steps of S4 are:

[0024] S401: Obtain identification information of the path disturbance affected segment, extract associated parameters of the affected path segment, calculate and record the offset of the path segment, determine whether conditions are met, and generate an offset analysis result of the disturbed path segment;

[0025] S402: Obtaining the propulsion direction and wind disturbance direction of the path segment, calculating the angle difference, determining whether it exceeds a preset deviation range, and generating an offset angle analysis value;

[0026] S403: Based on the angle offset analysis results, the path segments with excessive offset are screened, key node positions are adjusted, node position information in the high-level recirculation area is supplemented, and a post-disturbance path point position adjustment list is generated.

[0027] As a further embodiment of the present invention, the method further comprises:

[0028] S5: Based on the path data in the post-disturbance path point position adjustment list, the number of task points and the path length are extracted, and the balanced distribution of task points between paths is determined. If the coordination standard is not met, the path segment structure is adjusted and the task point position index is updated to establish the optimized path coordination distribution structure.

[0029] The optimized path coordination distribution structure includes the distribution of the number of task points, path length parameters, and a collaborative configuration structure.

[0030] As a further solution of the present invention, the specific steps of S5 are:

[0031] S501: Obtain the disturbed path point location data, extract the number of task points and the distance between task points on each path, obtain the number of task points and length parameters of the path, detect the distribution status of the task points, and generate path task point distribution data;

[0032] S502: Based on the task point distribution status data of the path, the distribution density of the task points is compared with a preset balance standard to screen out paths with uneven task point distribution, calculate the offset of the path, and generate path balance deviation data;

[0033] S503: Based on the path balance deviation data, the unbalanced path is adjusted, the path segment structure is reorganized, and the task point position index is adjusted, and the optimized multi-machine collaborative configuration result is output to obtain the optimized path coordination distribution structure.

[0034] A smart park drone inspection path optimization device, which is used to execute the smart park drone inspection path optimization method, includes:

[0035] The task point identification module is used to perform S1: obtain the three-dimensional coordinate information of the task points in the inspection area, identify the distance change trend of the path segments between the task points, calculate the average spacing of the task points in the path segment, compare it with the preset spacing threshold, filter out the path segments that exceed the set threshold, and obtain the path segment density identification result;

[0036] The path density analysis module is used to perform S2: based on the path segment density identification results, extract the task point set and calculate the center of gravity position of the set, determine the positional relationship between the set and the boundary points of the task segment, adjust the path connection structure, form a preliminary division of the collaborative paths of multiple UAVs, and obtain the inspection task allocation path set;

[0037] The flight path interference monitoring module is used to execute S3: extracting flight attitude data of inertial monitoring points in the high-temperature zone path section of the thermal pipeline in the plant area, identifying abnormal changes within the monitoring period, analyzing the wind disturbance direction trend, identifying the interfering path section and the disturbed direction, and generating path disturbance impact section identification information;

[0038] The path adjustment optimization module is used to execute S4: extract the affected path segments according to the path disturbance affected segment identification information, analyze the angular relationship between the propulsion direction and the wind disturbance direction, determine whether the offset exceeds the deviation range, adjust the path node positions, supplement the key node information, and obtain a list of path point position adjustments after the disturbance;

[0039] The multi-machine collaborative path allocation module is used to execute S5: based on the post-disturbance path point position adjustment list, extract the number and length parameters of the task points of the path, analyze the distribution status of the task points, and if it is unbalanced, reorganize the path structure and adjust the task point position index to obtain the optimized path coordination distribution structure.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are:

[0041] In the present invention, by obtaining the three-dimensional coordinates of the task points and analyzing the path spacing trend, dense sections can be identified and labeled, the spatial distribution structure perception is enhanced, the center of gravity and boundary relationship judgment are combined to realize task set screening, the independence of path division is guaranteed, the flight attitude data is integrated to identify the disturbance trend, the wind disturbance impact area is located, the offset node is adjusted and the key position information is supplemented, the path stability is optimized, the path parameters are extracted to analyze the distribution balance of the task points, the path structure is dynamically reconstructed, and multi-path load coordination is realized. The overall processing logic constitutes a closed loop of data perception, structure identification, interference analysis and scheduling optimization, thereby improving flight efficiency and path adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0043] Figure 2 A detailed flow chart of step S1 of the present invention;

[0044] Figure 3 A detailed flow chart of step S2 of the present invention;

[0045] Figure 4 A detailed flow chart of step S3 of the present invention;

[0046] Figure 5 A detailed flow chart of step S4 of the present invention;

[0047] Figure 6 The flowchart of step S5 of the present invention is refined;

[0048] Figure 7 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0051] Example 1

[0052] See also Figure 1 The present invention provides a technical solution: a method for optimizing the inspection path of a smart park drone, comprising the following steps:

[0053] S1: Obtain the three-dimensional coordinate information of the patrol task points in the closed area channel, identify the distance change trend between the task points in the continuous path segment, extract the average spacing value between the task points in the path segment, and compare it with the preset spacing threshold. Identify the path segment areas where the distance change amplitude exceeds the threshold setting interval, complete the area label division, and establish the path segment density recognition result;

[0054] S2: Extract the task point set based on the path segment density identification results, obtain the center of gravity of the set, and judge the relationship with the position of the existing task segment boundary points. Based on the judgment results, perform task set independence screening, adjust the path connection structure between the substation fence and the high-voltage line tower area, form the initial division of the multi-UAV collaborative path, and establish the inspection task allocation path set;

[0055] S3: Call the path segment of the high-temperature area of ​​the thermal pipeline in the factory area from the inspection task allocation path set, extract the flight attitude correlation data of the inertial monitoring points in the path, identify the abnormal change segments within the continuous monitoring cycle, mark the trend of wind disturbance direction changes, summarize the location and interference direction of the disturbed path segment, and establish the identification information of the path disturbance affected segment;

[0056] S4: Extract the affected path segments based on the path disturbance impact segment identification information, analyze the angular relationship between the propulsion direction and the wind disturbance direction, and determine whether the offset value exceeds the deviation range. If the conditions are met, uniformly adjust the node positions within the path segment, supplement the key node position information in the high-rise return flow area of ​​the building entrance and exit, and establish a list of path point position adjustments after the disturbance;

[0057] S5: Based on the path data of the drones in the list of path point positions after disturbance, the number and length parameters of the task points of each path are extracted to determine whether the distribution of task points between paths is balanced. If the coordination conditions are not met, the path segment structure is reorganized, the task point position index is adjusted, and the multi-machine collaborative configuration result is output to establish the optimized path coordination distribution structure.

[0058] The results of path segment density identification include the amplitude of the change in task point spacing, the area of ​​path segments exceeding the threshold, and the area label information. The inspection task allocation path set includes the independence screening task point set, the path connection structure, and the initial path division results. The path disturbance affected section identification information includes the abnormal change section location, wind disturbance direction trend, and the summary information of the disturbed path segment. The list of path point position adjustments after disturbance includes the node position adjustment value, key node position information, and the adjusted path node set. The optimized path coordination distribution structure includes the task point quantity distribution, path length parameters, and collaborative configuration structure.

[0059] See also Figure 2 , the specific steps of S1 are:

[0060] S101: Obtain the three-dimensional coordinate information of the inspection task points, construct a path segment, and sequentially extract the three-dimensional coordinate values ​​between adjacent task points. Calculate the distance between the task points based on the three-dimensional coordinate values ​​to generate the path segment distance value between the task points.

[0061] By acquiring the three-dimensional coordinates of patrol task points, the spatial location of each task point is determined. Task point coordinates, including X, Y, and Z dimensions, can be collected using GPS, drones, or 3D scanning technology. Using this coordinate information, path segments can be constructed between task points. Each path segment is a straight line connecting adjacent task points. When constructing a path segment, the three-dimensional distance between each two adjacent task points must first be calculated. This can be achieved by calculating the Euclidean distance between the two points. This distance calculation takes into account the differences in the X, Y, and Z coordinate axes. For example, if task point A has coordinates (3, 4, 5) and task point B has coordinates (6, 8, 9), the distance between them will be 3 units in X, 4 units in Y, and 4 units in Z, resulting in a final distance of approximately 6.4 units between task points A and B. The result of each calculation serves as the distance value for the path segment and is used for subsequent analysis and judgment.

[0062] S102: Calling the path segment distance value between task points, calculating the average spacing value between task points in the path segment, and performing a difference judgment with the set spacing threshold, extracting the path segment number and task point number that exceed the threshold, and obtaining the path segment spacing exceeding limit number value;

[0063] After obtaining the path segment distance values ​​between task points, the next step is to calculate the average spacing of all task points in the path segment. This is calculated by adding the distances between all path segments and then dividing by the total number of path segments. Assuming a path segment consists of multiple task points, and the distances between these path segments are 6.4, 7.2, 5.5, and 8.0 units, respectively, the average spacing between these path segments is 6.775 units. Next, by comparing with the set spacing threshold, it is determined which path segments have spacings exceeding the threshold. Assuming the set threshold is 6 units, when the distance of a path segment is greater than 6 units, the spacing of the path segment is considered to be outside the predetermined range. For example, if the distance of a path segment is 7.2 units, the spacing of this path segment will be marked as an out-of-limit path segment. In this step, the out-of-limit path segments will be extracted, and the task point numbers corresponding to these out-of-limit path segments will also be recorded to assist in subsequent density analysis and path optimization.

[0064] S103: Based on the path segment spacing overrun number value, adjacent overrun path segments are combined, the start and end task point numbers are extracted, the path segment area boundaries are constructed and annotated, and a path segment density recognition result is established;

[0065] Next, it is necessary to combine the adjacent out-of-limit path segments. Through this combination, dense areas can be identified. In this process, each out-of-limit path segment needs to be analyzed first to find which out-of-limit path segments are adjacent and combine them together to form a dense area. For example, if the distances of path segments C and D are both out of limit and they are adjacent, then these two path segments will be considered as a dense area. In this dense area, it is necessary to extract the numbers of the starting task point and the ending task point and mark the start and end boundaries of the area. For example, assuming that the starting task point of path segment C is numbered T3 and the ending task point is numbered T4, and the starting task point of path segment D is numbered T4 and the ending task point is numbered T5, then the task points in this dense area are numbered T3 to T5. Ultimately, these dense areas will be marked out and a density recognition result between the task points will be formed. In this way, areas with uneven spacing between task points can be clearly displayed, facilitating further path adjustment and optimization.

[0066] See also Figure 3 , the specific steps of S2 are:

[0067] S201: Obtain a trajectory node set in the path segment density recognition result, use node coordinate values ​​to calculate the centroid position of the path segment in the set, and generate a path segment centroid position coordinate set;

[0068] After obtaining the trajectory node set in the path segment density recognition result, the node set is first processed to generate the path segment centroid position coordinate set. First, for each node on the path segment, its corresponding coordinate value is obtained. A path segment consists of multiple nodes, so it is necessary to calculate the centroid position of each path segment. The specific operation is to sum the coordinate values ​​of all nodes in the path segment, and then obtain the centroid position of the path segment by averaging. In practical applications, this calculation can be assisted by map data or geographic information system (GIS) tools. Assuming that the node coordinates of a path segment are (1, 2), (2, 3), (3, 4), the centroid coordinates of the path segment are the arithmetic mean (2, 3) of the node coordinates, and this centroid point represents the center position of the path segment. By performing this operation on each path segment, the centroid coordinate set of all path segments can be obtained, providing a basis for the next step of calculation.

[0069] S202: Based on the coordinate set of the center of gravity position of the path segment, extract the spatial distance between the center of gravity point and the boundary point of the task segment, determine whether the center of gravity point is outside the boundary distance threshold range, and obtain a set of independent task path segments;

[0070] Based on the coordinate set of the path segment's centroid position, it is necessary to extract the spatial distance between the centroid and the task segment's boundary points and determine whether the centroid lies outside a threshold range. For each path segment's centroid, compare it with the corresponding task segment's boundary point. The distance between the centroid and the boundary point can be calculated using a simple distance calculation. The calculated distance is then compared with a pre-set threshold range. If the distance between the centroid and the boundary point is greater than the threshold, the path segment's centroid lies outside the threshold range. In this case, the path segment is considered an independent task path segment. For example, if the coordinates of the path segment's centroid are (2, 3) and the coordinates of the task segment's boundary point are (5, 7), the calculated spatial distance between them is 5. When the preset threshold is 3, the path segment is considered an independent task path segment. This type of calculation can be performed using a GIS platform. In practice, a script or program can be used to calculate the distance between the path segment and the boundary point and determine whether the path segment is an independent task path segment based on the threshold range.

[0071] S203: Based on the path connection relationship between the independent task path segment set and the fence area and the high-voltage line tower area, the connection structure between the path segment and the regional nodes is detected, a multi-path segment connection matrix is ​​established, and a multi-UAV collaborative path segment set is generated;

[0072] The specific calculation formula for the connection structure between path segments and regional nodes is:

[0073]

[0074] Calculate the connection matrix between path segments and regional nodes to generate a set of collaborative path segments for multiple UAVs;

[0075] Among them, C ij represents the connectivity between path segment i and regional node j, P ik represents the coordinates of path segment i in the kth dimension, P jk represents the coordinate of region node j in the kth dimension, d ij is the distance between path segment i and regional node j, w m is the weight of the mth path segment, n is the number of dimensions of the path segment, and N is the total number of path segments;

[0076] The parameters in the formula are:

[0077] C ij is the connectivity between path segment i and regional node j, indicating the relative closeness between path segment i and regional node j;

[0078] P ik is the coordinate of path segment i in the kth dimension, which is obtained by monitoring the actual position data of the path segment; for example, if the three coordinate dimensions of path segment i are (10, 20, 30), then P i1 =10, P i2 =20, P i3 =30;

[0079] P jk is the coordinate of regional node j in the kth dimension, which is similarly obtained through data collection of regional nodes;

[0080] For example, if the three coordinate dimensions of region node j are (12, 18, 28), then P j1 =12,P j2 =18, P j3 =28;

[0081] d ij is the distance between path segment i and regional node j, which is usually calculated using the Euclidean distance formula:

[0082]

[0083] By substituting specific values, we get:

[0084]

[0085] w mis the weight of path segment m, which is usually quantified by the importance or coverage of the path segment; weight w m It can be allocated based on the criticality of the drone’s path;

[0086] Assume that in a certain scenario, the importance of path segments is distributed by pre-set coefficients, for example, the weight of path segment m is 1.2;

[0087] n is the dimension of the path segment, usually the dimension of the space where the path segment is located;

[0088] For example, if the coordinates of the path segments are in three-dimensional space, n = 3;

[0089] N is the total number of path segments, which is the number of all path segments. Assume that in this task, there are 10 path segments, N = 10;

[0090] The calculation process of the derived formula is as follows: 1. Calculate the sum of the squares of the coordinate differences between path segment i and regional node j:

[0091]

[0092] Perform a square root operation on the result to obtain the distance metric between path segment i and region node j:

[0093]

[0094] Calculate the Euclidean distance between path segment i and region node j:

[0095] d ij =3.464;

[0096] Calculate the sum of the weights of all path segments. Assuming the weights of the path segments are 1.2, 0.8, 1.5, 1.1, 0.9, 1.3, 1.0, 1.6, 1.2, and 0.7, the sum of the path segment weights is:

[0097]

[0098] Substitute all calculated results into the original formula:

[0099]

[0100] Result interpretation: This result indicates that the connectivity between path segment i and regional node j is 0.0884. A low connectivity value indicates that the connection between path segment i and regional node j is relatively distant, and the path may need to be adjusted or the UAV collaborative path planning may need to be further optimized.

[0101] The calculation logic in the formula is first calculated based on the geometric distance and relative position relationship between the path segment and the regional node. First, the sum of the squares of the coordinate differences in each dimension between the path segment i and the regional node j is calculated. This operation captures the spatial differences between the dimensions by subtracting and squaring. The purpose is to measure the deviation between the path segment and the node in each dimension. Then, the square sum is converted into the actual Euclidean distance by taking the square root. This step is to restore the square sum of the coordinate differences to the original physical space scale, so that the calculation result reflects the actual spatial distance, not just the sum of the squares of the differences. Subsequently, this distance value is divided by the Euclidean distance d between the path segment and the regional node. ij , this process normalizes the distance value into a unit distance metric to prevent large deviations in the value due to spatial differences. Finally, the weight factor of the path segment is introduced by multiplying it by the inverse of the sum of the path segment weights. This step allows path segments with different weights to reflect their relative importance in the calculation of connectivity. Path segments with high weights will have a greater impact on the connectivity calculation, thereby adjusting the accuracy of path planning and the flexibility of UAV collaborative paths. Overall, the formula comprehensively measures the connection strength between path segments and regional nodes through the calculation of spatial distance and the introduction of path weights;

[0102] The connection matrix between path segments and regional nodes is used to describe the relative connectivity and closeness between path segments and regional nodes. It calculates the connectivity between each path segment and each regional node, forming a matrix structure in which each element represents the strength of the connection between a specific path segment and a specific regional node. The significance of this matrix lies in providing a quantitative method to help analyze and evaluate the spatial correlation between path segments and regional nodes, thereby providing data support for tasks such as path optimization, drone collaborative operations, and regional monitoring. This matrix can be used to identify which path segments have closer connections with specific regional nodes, which is important for drone scheduling, path planning, and the rational allocation of collaborative operations.

[0103] See also Figure 4 , the specific steps of S3 are:

[0104] S301: Acquire inertial monitoring point data in a set of collaborative path segments of multiple UAVs, combine the flight attitude information of each UAV, integrate and analyze the data of each monitoring point in the path segment, and generate flight attitude data;

[0105] First, inertial monitoring point data is collected from multiple drones. Each drone is equipped with an inertial measurement unit (IMU) to collect information such as acceleration and angular velocity in real time, and combined with the position information provided by the GPS module to obtain flight attitude data. The flight attitude can be calculated by calculating the acceleration and angular velocity to obtain the pitch angle, roll angle, heading angle, etc. of the drone. Assume that the attitude of the drone in flight is a pitch angle of 10°, a roll angle of 5°, and a heading angle of 45°, and at a certain monitoring point, the recorded acceleration is 0.03m / s 2 Using this data, we can further analyze each monitoring point in the path segment, integrate the flight data of each drone, and generate comprehensive flight attitude data for subsequent processing and analysis.

[0106] S302: Based on the flight attitude data, abnormal fluctuations within the monitoring period are determined using a set threshold, abnormal change segments in the path segments are identified, and abnormal change identifiers are generated;

[0107] Based on the flight attitude data, a threshold is set to determine whether abnormal fluctuations occur during flight. The threshold is generally derived from flight tests or historical data. For example, the fluctuation range of the pitch angle, roll angle, and heading angle is set. If the attitude angle change of a certain monitoring point exceeds the set threshold, such as the pitch angle fluctuation exceeds 3°, the roll angle fluctuation exceeds 2°, and the heading angle fluctuation exceeds 5°, it is determined to be an abnormal fluctuation. These abnormal fluctuations can be used to identify abnormal changes in path segments. When the fluctuation of a monitoring point on a certain path segment exceeds the threshold, the path segment is marked as an abnormal segment, and the time point and corresponding location information of the abnormality are recorded to facilitate subsequent further analysis. For example, if the pitch angle change of a certain monitoring point exceeds 3°, it is marked as an abnormal segment, which may affect flight safety and require subsequent intervention or adjustment.

[0108] S303: Analyze the changing trends of the disturbed path segments and wind disturbance directions based on the abnormal change identification, mark the positions and directions of the disturbed path segments, and generate identification information of the path disturbance affected sections;

[0109] Based on the abnormal change indicators, the changing trends of disturbed path segments and the impact of wind disturbance direction can be further analyzed. By obtaining wind speed and direction data during flight and combining it with flight attitude information, the impact of wind can be analyzed. For example, if the wind speed is high and the angle between the wind direction and the flight path direction exceeds 30°, the wind is considered to have a significant impact on the flight path. Assuming a wind speed of 15m / s and a northwest wind direction, deviation from the flight path has a significant impact. In the path segment analysis, the deviation between the planned flight trajectory and the actual flight trajectory is first compared to calculate the path offset. When the offset reaches a certain standard, the path segment is considered to be significantly disturbed by wind. Assuming that the actual trajectory deviates from the planned trajectory by 20m, the path segment is marked as a wind disturbance segment, and the start and end times of the wind disturbance, as well as the change in wind direction, are recorded. Marking the location and direction of the disturbed path segment helps flight management personnel adjust or optimize the path segment to ensure flight safety and reduce the impact of disturbances.

[0110] See also Figure 5 , the specific steps of S4 are:

[0111] S401: Obtain identification information of the path disturbance affected segment, extract associated parameters of the affected path segment, calculate and record the offset of the path segment, determine whether the conditions are met, and generate an offset analysis result of the disturbed path segment;

[0112] The system obtains the identification information of the affected path segment, extracts relevant parameters of the affected path segment, calculates and records the path segment's deviation, determines whether the conditions are met, and generates a deviation analysis result for the disturbed path segment. First, the identification information of the path segment affected by the path disturbance is obtained using positioning technology. The affected area is determined by combining the path segment's geographic coordinates, operating status, climate data, and other data. Specifically, when natural factors such as wind speed and temperature fluctuate, these factors will cause certain parts of the path segment to be disturbed. The affected area is determined using input data from sensors and monitoring systems. Next, relevant parameters of the affected path segment are extracted, including its length, material, location, and load capacity. These parameters are automatically collected by sensor equipment and processed by a data analysis system. When calculating deviation, the physical condition of the path segment is analyzed using historical records, real-time data, and equipment health status to record the specific path deviation. If a path segment physically deviates, such as bending or displacement due to wind, the system records the extent and specific location of the deviation. The path segment's offset is then compared to a preset tolerance range. If a path segment's offset exceeds the tolerance (for example, an offset exceeding 5 meters), the segment is considered to require repair. Finally, the system generates an offset analysis, listing the offset data and repair recommendations for each affected path segment, providing a basis for subsequent work.

[0113] S402: Obtain the propulsion direction and wind disturbance direction of the path segment, calculate the angle difference, determine whether it exceeds a preset deviation range, and generate an offset angle analysis value;

[0114] The propulsion direction and wind disturbance direction of the path segment are obtained, the angle difference is calculated, and it is determined whether it exceeds the preset deviation range to generate an offset angle analysis value. First, the propulsion direction of the path segment is determined by the data acquisition equipment. The actual propulsion direction of the path is obtained by combining the travel trajectory of the path segment and its actual use (such as the direction of the power transmission line or the transportation route). At the same time, the wind disturbance direction is obtained by the wind direction monitoring device, which records the changes in wind direction and real-time data of wind speed. Then, the angle difference between the propulsion direction of the path segment and the wind disturbance direction is calculated. The angle difference refers to the deviation angle between the propulsion direction of the path segment and the wind direction. If there is a large angle difference between the wind direction and the propulsion direction of the path segment, the stability of the path may be affected. At this time, by comparing the angles between the two, if the angle difference exceeds the preset tolerance range (for example, 15°), it is considered that the angle deviation of the path segment exceeds the standard. Based on this analysis result, an offset angle analysis value is generated, which is used for subsequent judgment and correction, providing a reference for path optimization.

[0115] S403: Based on the angle offset analysis results, the path segments with excessive offset are screened out, the key node positions are adjusted, the node position information in the high-level recirculation area is supplemented, and a list of adjusted path point positions after the disturbance is generated;

[0116] The specific calculation formula for adjusting the position of key nodes in the path segment is:

[0117]

[0118] Calculate the offset of key nodes, the adjustment of path segments, and the reflux information of key nodes;

[0119] Where ΔP represents the total adjustment of the key node position of the path segment, represents the new position of the i-th path point after disturbance adjustment, represents the original position of the i-th path point, d i Indicates the distance between the i-th path point and the center of the adjustment area, in meters (m), r i Indicates the return radius of the i-th path point, in meters (m), Indicates the location of the j-th high-level recirculation area node, in meters (m), represents the position of the jth basic node in meters (m), n is the total number of nodes in the path segment, and m is the number of nodes in the reflow area;

[0120] The new position of the i-th path point after disturbance adjustment, in meters. The geographical coordinate changes of the path point are obtained in real time through monitoring equipment, and the new position coordinates are calculated;

[0121] The original position of the i-th waypoint, in meters. The original geographic coordinates of the waypoint are obtained through historical data or initial planning;

[0122] d i : The distance between the i-th path point and the center of the adjustment area, in meters. The straight-line distance between the path point and the center of the adjustment area is calculated using a geographic information system (GIS);

[0123] r i : The return flow radius of the i-th path point, in meters. The return flow influence range of the path point is determined through fluid dynamics models or historical data analysis;

[0124] The location of the j-th high-rise recirculation area node, in meters. The location of the high-rise recirculation area node is determined by combining the height and location data of high-rise buildings with the wind speed and direction model;

[0125] The location of the jth infrastructure node, in meters. Obtain the geographic coordinates of the infrastructure node from the infrastructure design drawings or construction data.

[0126] Parameter value setting basis:

[0127] and Obtain data based on actual monitoring data and historical planning data to ensure data accuracy and timeliness;

[0128] d i : Calculate the distance between the path point and the center of the adjustment area through the GIS system to ensure the accuracy of the calculation results;

[0129] r i : Based on the fluid mechanics model or historical data analysis, determine the scope of the backflow influence of the path point to ensure the rationality of the model;

[0130] and Obtain node locations based on building height and location data, as well as infrastructure design drawings, to ensure data accuracy;

[0131] Formula calculation derivation process:

[0132] Assume there are 3 path points and 2 return flow area nodes, the specific data is as follows:

[0133] Waypoint data:

[0134]

[0135] Recirculation area node data:

[0136]

[0137] Calculate the waypoint adjustment:

[0138] 0.0068;

[0139] Total waypoint adjustment:

[0140] ΔP total =0.0094+0.0079+0.0068=0.0241;

[0141] Calculate the node adjustment amount in the recirculation area:

[0142] ΔP high =|105.0-104.5|+|106.0-105.5|=0.5+0.5=1.0;

[0143] The final path segment key node position adjustment amount:

[0144] ΔP=0.0241+1.0=1.0241;

[0145] Result interpretation:

[0146] The results show that the total adjustment of the key node positions of the path segment is 1.0241 meters. This value reflects the displacement of the path points after the disturbance adjustment and the degree of influence of the nodes in the recirculation area on the path adjustment.

[0147] The formula's operational logic is based on the actual needs of adjusting the positions of key nodes in a path segment. The use of operators such as addition, multiplication, and square root between parameters reflects the impact of different factors on the path adjustment. First, the formula calculates the difference between the new and original positions of each path point to obtain the path point adjustment. This difference is related to the path point's distance from the center of the adjustment area and its recirculation radius. The relationship between distance and recirculation radius is manipulated using square roots to reflect the influence of the recirculation area. Specifically, the distance between the path point and the center of the adjustment area is combined through addition, ensuring that the size of the adjustment range influences the path point's adjustment amplitude. The introduction of square roots reflects the nonlinear effect of the recirculation radius on the path point adjustment: that is, when the recirculation radius is larger, the path point adjustment changes less. The adjustments of all path points are summed to obtain the total path segment adjustment. Furthermore, the adjustments of nodes in the recirculation area are directly calculated by adding the differences in their positions, reflecting the contribution of nodes within the recirculation area to the path adjustment. Ultimately, the total path segment adjustment is the sum of the path point adjustments and the recirculation area node adjustments, integrating the combined influence of all factors on the path adjustment.

[0148] "Key node offset" refers to the displacement of the positions of key nodes in a path segment due to disturbance or adjustment, reflecting the relative change in each node on the path after adjustment. This offset can reveal which node positions have undergone significant changes during the path optimization process, thereby affecting the overall shape and performance of the path. "Path segment adjustment" refers to the sum of the adjustments made to all path points, reflecting the overall degree of adjustment experienced by the entire path segment during the optimization process. It is an indicator that quantifies the overall effect of path optimization, reflecting the overall magnitude of change from the original plan to the final adjustment after path optimization. "Key node reflow information" refers to node position information related to high-level reflow areas, showing changes in node positions within the reflow-affected area. Reflow information helps further analyze the performance of path adjustments in high-level areas, ensuring that the special requirements and impacts of high-level reflow areas are fully considered during the path optimization process, thereby improving the overall accuracy and practicality of the path.

[0149] See also Figure 6 , the specific steps of S5 are:

[0150] S501: Obtain the disturbed path point location data, extract the number of task points and the distance between task points on each path, obtain the number of task points and length parameters of the path, detect the distribution status of the task points, and generate the path task point distribution data;

[0151] After perturbing the path point location data, the number of task points on each path and the distances between adjacent task points are first extracted. Path point perturbation can be achieved by simulating environmental interference, such as adding a certain amount of noise, causing the points in the original path to shift slightly. The perturbed path points provide a different task point distribution for subsequent analysis. Next, by counting the number of task points on each path, the total number of task points on each path is determined, and the distances between adjacent task points are further calculated. For example, if the distance between task points 1 and 2 on a path is 2.5 meters, the distance between task points 2 and 3 is 3.2 meters, and so on, the distances between these task points can be accumulated to obtain the total length of the path. The distribution of these task points is then analyzed to determine whether the task points on the path are evenly spaced and whether there are any overcrowding or sparseness. Based on this information, path task point distribution data is generated, reflecting the distribution of the path task points and providing basic data for subsequent path optimization.

[0152] S502: Based on the task point distribution status data of the path, the distribution density of the task points is compared with the preset balance standard to judge, and the paths with uneven task point distribution are screened out, the offset of the path is calculated, and the path balance deviation data is generated;

[0153] Based on the distribution data of task points along a path, the density of task points can be calculated using the proportional relationship between the path length and the number of task points. If the path length is L and the total number of task points is n, the density of task points along the path can be derived using a simple proportional relationship. The preset balance standard is a target task point density range, which may be determined by empirical data or actual requirements. For example, the target density may be set at 0.2 to 0.5 task points per meter of path. When the actual density of a path exceeds this range, the task point distribution is considered unbalanced. Calculating the path offset is a key step, quantifying the degree of deviation from the path distribution. The offset is calculated by comparing the actual distribution density with the target equilibrium density. When the actual density of a path is significantly lower than the target density, the offset is larger, and vice versa. By calculating the offset for each path, paths with unbalanced task point distribution are identified, and path balance deviation data is generated to facilitate subsequent path adjustment and optimization.

[0154] S503: Based on the path balance deviation data, the unbalanced path is adjusted, the path segment structure is reorganized, and the task point location index is adjusted. The optimized multi-machine collaborative configuration result is output to obtain the optimized path coordination distribution structure.

[0155] When adjusting unbalanced paths, the path offset is first assessed using path balance deviation data. If a path's offset exceeds a set threshold (for example, an offset exceeding 0.1), the path is considered in need of optimization. Next, the path segment structure is reorganized, adjusting the distribution of task points across segments. Each path can be re-divided into several more evenly spaced segments based on the desired task point distribution, with the number of task points within each segment approaching consistency. If a segment has fewer task points, the distribution can be adjusted by inserting new task points or increasing the length of the path segment. If a segment has more task points, balancing can be achieved by reducing the length of the path segment. When adjusting task point positions, the index order of the path points may need to be rearranged to achieve a more even distribution of task points. Through this series of adjustments, the optimized path will have a more balanced distribution of task points, avoiding the previously existing situation of dense or sparse task points, ensuring a more balanced and efficient execution of multi-machine collaborative tasks.

[0156] See also Figure 7 , a smart park drone inspection path optimization device, comprising:

[0157] The task point recognition module obtains the three-dimensional coordinate information of the task points in the inspection area, identifies the distance change trend of the path segments between the task points, calculates the average spacing of the task points in the path segment, compares it with the preset spacing threshold, and filters the path segments that exceed the set threshold to obtain the path segment density recognition result;

[0158] The path density analysis module extracts the task point set and calculates the center of gravity of the set based on the path segment density identification results. It determines the positional relationship between the set and the boundary points of the task segment, adjusts the path connection structure, forms a preliminary division of the collaborative paths of multiple UAVs, and obtains the inspection task allocation path set.

[0159] The flight path interference monitoring module extracts flight attitude data from inertial monitoring points along the high-temperature path sections of the plant's thermal pipelines, identifies abnormal changes within the monitoring period, analyzes wind disturbance direction trends, identifies interfering path sections and disturbed directions, and generates identification information for path disturbance-affected sections.

[0160] The path adjustment and optimization module extracts the affected path segments based on the identification information of the path disturbance affected sections, analyzes the angular relationship between the propulsion direction and the wind disturbance direction, determines whether the offset exceeds the deviation range, adjusts the path node positions, supplements key node information, and obtains a list of path point position adjustments after the disturbance;

[0161] The multi-machine collaborative path allocation module extracts the number and length parameters of the task points of the path based on the post-disturbance path point position adjustment list, analyzes the distribution status of the task points, and if there is imbalance, reorganizes the path structure and adjusts the task point position index to obtain the optimized path coordination distribution structure.

[0162] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for optimizing the inspection path of a smart park drone, characterized in that: The following steps are involved: S1: Obtain the three-dimensional coordinates of the patrol task points in the closed area channel, identify the distance change trend between the task points in the path segment, extract the average spacing value, compare it with the preset spacing threshold, determine the path segment area where the change range exceeds the threshold interval, and establish the path segment density recognition result; S2: extracting a task point set based on the density identification result of the path segment, determining the positional relationship between the set center and the task segment boundary, screening the independence of the task set, adjusting the path structure of the substation fence and high-voltage line tower area, and establishing an inspection path allocation set; S3: Calling the high-temperature area path segment of the thermal pipeline in the plant area from the inspection task allocation path set, extracting the flight attitude correlation data of the inertial monitoring points in the path, identifying and marking the path segment with continuous changes within the monitoring period, and establishing the path disturbance impact segment identification information; S4: Extract the affected path segment based on the path disturbance impact segment identification information, analyze the angular relationship between the propulsion direction and the wind disturbance direction, and determine whether the offset value exceeds the deviation range. If the conditions are met, uniformly adjust the node positions within the path segment and establish a list of path point position adjustments after the disturbance.

2. The smart park drone inspection path optimization method according to claim 1 is characterized in that: The path segment density identification result includes the task point spacing change amplitude, the path segment area exceeding the threshold, and the area label information; the inspection task allocation path set includes the independence screening task point set, the path connection structure, and the initial path division result; the path disturbance impact section identification information includes the abnormal change section position, the wind disturbance direction trend, and the disturbed path segment summary information; the post-disturbance path point position adjustment list includes the node position adjustment value, the key node position information, and the adjusted path node set.

3. The method for optimizing the inspection path of a smart park drone according to claim 1 is characterized in that: The specific steps of S1 are: S101: Obtain the three-dimensional coordinate information of the inspection task points, construct a path segment, and sequentially extract the three-dimensional coordinate values ​​between adjacent task points. Calculate the distance between the task points based on the three-dimensional coordinate values ​​to generate the path segment distance value between the task points. S102: Calling the path segment distance value between the task points, calculating the average spacing value between the task points in the path segment, and performing a difference judgment with the set spacing threshold, extracting the path segment number and task point number that exceed the threshold, and obtaining the path segment spacing exceeding limit number value; S103: Based on the path segment spacing overrun number value, adjacent overrun path segments are combined, start and end task point numbers are extracted, path segment area boundaries are constructed and annotated, and a path segment density recognition result is established.

4. The method for optimizing the inspection path of a smart park drone according to claim 3 is characterized in that: The specific steps of S2 are: S201: Obtain a trajectory node set in the path segment density recognition result, calculate the centroid position of the path segment in the set using the node coordinate values, and generate a path segment centroid position coordinate set; S202: Based on the coordinate set of the center of gravity position of the path segment, extract the spatial distance between the center of gravity point and the boundary point of the task segment, determine whether the center of gravity point is outside the boundary distance threshold range, and obtain a set of independent task path segments; S203: According to the path connection relationship between the independent task path segment set and the fence area and the high-voltage line tower area, the connection structure between the path segment and the regional node is detected, a multi-path segment connection matrix is ​​established, and a multi-UAV collaborative path segment set is generated.

5. The method for optimizing the inspection path of a smart park drone according to claim 4 is characterized in that: The specific steps of S3 are: S301: Acquire inertial monitoring point data in the collaborative path segment set of the multiple UAVs, combine the flight attitude information of each UAV, integrate and analyze the data of each monitoring point in the path segment, and generate flight attitude data; S302: Based on the flight attitude data, determine abnormal fluctuations within the monitoring period using a set threshold, identify abnormal change segments in the path segments, and generate abnormal change identifiers; S303: Analyze the changing trends of the disturbed path segments and wind disturbance directions according to the abnormal change identifiers, mark the positions and directions of the disturbed path segments, and generate path disturbance impact segment identifier information.

6. The method for optimizing the inspection path of a smart park drone according to claim 5 is characterized in that: The specific steps of S4 are: S401: Obtain identification information of the path disturbance affected segment, extract associated parameters of the affected path segment, calculate and record the offset of the path segment, determine whether conditions are met, and generate an offset analysis result of the disturbed path segment; S402: Obtaining the propulsion direction and wind disturbance direction of the path segment, calculating the angle difference, determining whether it exceeds a preset deviation range, and generating an offset angle analysis value; S403: Based on the angle offset analysis results, the path segments with excessive offset are screened, key node positions are adjusted, node position information in the high-level recirculation area is supplemented, and a post-disturbance path point position adjustment list is generated.

7. The method for optimizing the inspection path of a smart park drone according to claim 1 is characterized in that: The method further comprises: S5: Based on the path data in the post-disturbance path point position adjustment list, the number of task points and the path length are extracted, and the balanced distribution of task points between paths is determined. If the coordination standard is not met, the path segment structure is adjusted and the task point position index is updated to establish the optimized path coordination distribution structure. The optimized path coordination distribution structure includes the distribution of the number of task points, path length parameters, and a collaborative configuration structure.

8. The method for optimizing the inspection path of a smart park drone according to claim 7 is characterized in that: The specific steps of S5 are: S501: Obtain the disturbed path point location data, extract the number of task points and the distance between task points on each path, obtain the number of task points and length parameters of the path, detect the distribution status of the task points, and generate path task point distribution data; S502: Based on the task point distribution status data of the path, the distribution density of the task points is compared with a preset balance standard to screen out paths with uneven task point distribution, calculate the offset of the path, and generate path balance deviation data; S503: Based on the path balance deviation data, the unbalanced path is adjusted, the path segment structure is reorganized, and the task point position index is adjusted, and the optimized multi-machine collaborative configuration result is output to obtain the optimized path coordination distribution structure.

9. A smart park drone inspection path optimization device, characterized in that: The smart park drone inspection path optimization device is used to execute the smart park drone inspection path optimization method according to any one of claims 1 to 8, and the smart park drone inspection path optimization device includes: The task point identification module is used to perform S1: obtain the three-dimensional coordinate information of the task points in the inspection area, identify the distance change trend of the path segments between the task points, calculate the average spacing of the task points in the path segment, compare it with the preset spacing threshold, filter out the path segments that exceed the set threshold, and obtain the path segment density identification result; The path density analysis module is used to perform S2: based on the path segment density identification results, extract the task point set and calculate the center of gravity position of the set, determine the positional relationship between the set and the boundary points of the task segment, adjust the path connection structure, form a preliminary division of the collaborative paths of multiple UAVs, and obtain the inspection task allocation path set; The flight path interference monitoring module is used to execute S3: extracting flight attitude data of inertial monitoring points in the high-temperature zone path section of the thermal pipeline in the plant area, identifying abnormal changes within the monitoring period, analyzing the wind disturbance direction trend, identifying the interfering path section and the disturbed direction, and generating path disturbance impact section identification information; The path adjustment optimization module is used to execute S4: extract the affected path segments according to the path disturbance affected segment identification information, analyze the angular relationship between the propulsion direction and the wind disturbance direction, determine whether the offset exceeds the deviation range, adjust the path node positions, supplement the key node information, and obtain a list of path point position adjustments after the disturbance; The multi-machine collaborative path allocation module is used to execute S5: based on the post-disturbance path point position adjustment list, extract the number and length parameters of the task points of the path, analyze the distribution status of the task points, and if it is unbalanced, reorganize the path structure and adjust the task point position index to obtain the optimized path coordination distribution structure.

10. A smart park drone inspection path optimization device, including a memory and a processor, characterized in that: A computer program is stored in the memory, and when the processor executes the computer program, the smart park drone inspection path optimization device according to claim 9 is implemented.

Images