A UAV multi-mode mine inspection method, system, medium and program product

Abstract

A multi-mode mine inspection method, system, medium and program product for unmanned aerial vehicles, in which the inspection targets are divided into short-range objects and long-range objects based on a preset inspection target classification list; a three-dimensional grid inspection path is constructed according to the location information of the short-range objects and the long-range objects, and a number of spare inspection entry points are set in the detection area of ​​the short-range objects; real-time environmental parameters are obtained, and the real-time environmental parameters include light intensity, atmospheric visibility, electromagnetic field intensity, and wind speed and direction; the accessibility score of each spare inspection entry point is calculated based on the degree of influence of the real-time environmental parameters on the unmanned aerial vehicle, and the spare inspection entry point with the highest accessibility score is determined as the current inspection entry point; an inspection path is generated according to the current inspection entry point, and the unmanned aerial vehicle is controlled to perform inspections along the inspection path. This application improves the integrity of the inspection data, improves the accuracy of the safety assessment data obtained based on the inspection data, and thus improves the safety of the mine.

Description

Technical Field

[0001] The present application belongs to the field of data processing methods specifically used for management, and in particular, relates to a drone multi-mode mine inspection method, system, medium and program product. Background Art

[0002] With the rapid development of drone technology, drones have been widely used in the field of mine inspection. Traditional mine inspection mainly relies on manual inspection, which has problems such as high workload, low efficiency and high risk factor. With its advantages of flexibility and mobility, drones can effectively solve the defects of traditional manual inspection and improve the efficiency and safety of mine inspection.

[0003] In related technologies, aerial photography of mining areas can be performed through preset flight paths to obtain image data of mining areas, and then the image data can be analyzed manually or automatically to identify abnormal conditions in mining areas, such as collapse, landslides, cracks, etc. This method uses the aerial photography capabilities of drones to improve the efficiency of mine inspections to a certain extent, reduce the workload of manual inspections, and achieve remote monitoring of mining areas and timely detection of abnormal conditions.

[0004] However, if drones need to inspect some special objects, such as high-voltage power facilities and precision instruments and equipment, they need to observe and detect them at close range. However, due to insufficient light and radio signal attenuation, it is still difficult to achieve comprehensive and effective coverage of all areas and targets in the complex and changeable mining environment, which reduces the integrity of the inspection data and the accuracy of the safety assessment data obtained based on the inspection data, thereby reducing the safety of the mine. Summary of the invention

[0005] The present application provides a drone multi-mode mine inspection method, system, medium and program product for improving the integrity of inspection data, improving the accuracy of safety assessment data obtained based on the inspection data, and thereby improving the safety of mines.

[0006] In a first aspect, the present application provides a multi-mode mine inspection method using a drone, which divides inspection targets into short-range objects and long-range objects based on a preset inspection target classification list, wherein short-range objects include inspection targets whose detection distance is less than a preset distance, and long-range objects include inspection targets that need to be observed as a whole;

[0007] Construct a three-dimensional grid inspection path based on the location information of the near-range object and the far-range object, and set a number of backup inspection entry points in the detection area of ​​the near-range object;

[0008] Obtain real-time environmental parameters, including light intensity, atmospheric visibility, electromagnetic field intensity, and wind speed and direction;

[0009] Calculate the accessibility score of each backup inspection entry point based on the degree of influence of real-time environmental parameters on the drone, and determine the backup inspection entry point with the highest accessibility score as the current inspection entry point;

[0010] Generate an inspection path based on the current inspection entry point, and control the drone to inspect along the inspection path.

[0011] By adopting the above technical solution, the inspection targets are divided into short-range objects and long-range objects. Combined with the setting of three-dimensional grid inspection paths and spare inspection entry points, differentiated inspection strategies can be formulated for inspection targets with different detection distances. By obtaining real-time environmental parameters such as light intensity, atmospheric visibility, electromagnetic field intensity, wind speed and direction, calculating the accessibility score of spare inspection entry points and selecting the optimal entry point, the drone can select the most suitable inspection path according to the actual environmental conditions, improving the reliability and efficiency of the inspection task, reducing the adverse effects of environmental factors on drone inspections, and at the same time enhancing the flexibility of the inspection path through the setting of spare entry points, so that the inspection system can better cope with the complex and changeable mining environment, improving the integrity of the inspection data, and improving the accuracy of the safety assessment data obtained based on the inspection data, thereby improving the safety of the mine.

[0012] Furthermore, the accessibility score of each backup inspection entry point is calculated based on the degree of influence of real-time environmental parameters on the drone, including:

[0013] Input the real-time environmental parameters into the impact degree calculation function to obtain the impact degree score;

[0014] A three-dimensional vector is constructed based on the distance from the current position of the drone to each backup inspection entry point and the terrain height difference, and the flight energy consumption coefficient is calculated based on the degree of influence;

[0015] Calculate the remaining power to each standby inspection entry point based on the flight energy consumption coefficient;

[0016] Calculate the navigation risk value based on the impact level and remaining power;

[0017] The accessibility of each alternative inspection entry point is evaluated based on the navigation risk value to obtain the accessibility benchmark score;

[0018] The accessibility benchmark score is weighted with the historical success rate of each backup inspection entry point to obtain the accessibility score.

[0019] By adopting the above technical solutions, an accessibility scoring mechanism including multi-dimensional indicators such as distance, terrain height difference, environmental impact, energy consumption and historical success rate is established, realizing a comprehensive evaluation of the backup inspection entry points. This scoring mechanism converts environmental parameters into specific impact scores, and calculates the flight energy consumption coefficient in combination with the drone's location information, further estimates the remaining power and navigation risk value, and finally obtains the accessibility score through weighted calculation with the historical success rate. This multi-level evaluation method can fully reflect the actual accessibility of each entry point, reduce the risk of mission failure due to poor environmental conditions or insufficient energy, and improve the accuracy and scientificity of entry point selection.

[0020] Furthermore, the influence degree calculation function is: In the function, To measure the degree of impact, is the total duration of the inspection task, is the time variable, is the reference value of light intensity, for The light intensity at the time, is the reference value of visibility, for Visibility at all times, is the standard deviation of visibility, for The electromagnetic field strength at the time, is the influence function of electromagnetic field intensity, for The wind speed at the moment, for The wind direction angle at the moment, is the joint influence function of wind speed and wind direction, is the weight coefficient of light intensity, is the adjustment coefficient within the logarithmic function of light intensity, is the weight coefficient of visibility, is the weight coefficient of electromagnetic field strength, is the weight coefficient of wind speed and direction.

[0021] By adopting the above technical solution, an integral function model including multiple environmental factors such as light intensity, visibility, electromagnetic field intensity, wind speed and direction is constructed to achieve accurate quantification of environmental impact. The function model uses a logarithmic function to describe the influence characteristics of light intensity, uses a Gaussian function to characterize the influence law of visibility, and uses an influence function to characterize the role of electromagnetic field intensity, while considering the combined influence of wind speed and wind direction. The weight coefficient introduced in the model allows the influence degree of each environmental factor to be flexibly adjusted, and the integral form ensures that the cumulative effect of environmental impact is fully considered during the entire inspection process.

[0022] Furthermore, after controlling the drone to perform inspection along the inspection path, the method further includes:

[0023] Based on the real-time environmental parameters, the detection equipment of the UAV is adjusted in parameters, and the parameter adjustment includes adjusting the working frequency and transmission power of the detection equipment to obtain dynamic compensation parameters;

[0024] Acquire the sampling data of real-time environmental parameters in real time, and calculate the change of environmental parameters between two adjacent sampling moments;

[0025] When the environmental parameter change exceeds a preset threshold, several candidate evacuation paths are calculated based on the backup inspection entry points;

[0026] The time cost and safety factor of several candidate evacuation paths are calculated, and the evacuation path with the lowest time cost and a safety factor that meets the preset conditions is selected as the optimal evacuation path.

[0027] By adopting the above technical solutions, a dynamic adjustment mechanism for detection equipment parameters and a real-time planning method for emergency evacuation routes are established, which enhances the environmental adaptability of the drone inspection system. The system can dynamically compensate the working frequency and transmission power of the detection equipment according to the environmental parameters, and timely calculate multiple candidate evacuation routes when environmental conditions deteriorate by monitoring the changing trends of environmental parameters in real time. By comprehensively considering the time cost and safety factor, the optimal evacuation route is selected, which not only improves the timeliness and safety of the evacuation process, but also improves the inspection system's ability to respond to sudden environmental changes and reduces the risk of equipment damage and mission failure.

[0028] Furthermore, the detection equipment of the UAV is adjusted based on the real-time environmental parameters, including:

[0029] Calculate the signal attenuation coefficient based on the electromagnetic field strength data in the real-time environmental parameters;

[0030] adjusting the transmission power of the detection device based on the signal attenuation coefficient;

[0031] Determine the signal transmission quality level according to the atmospheric visibility data in the real-time environmental parameters;

[0032] The operating frequency of the detection device is adjusted based on the signal transmission quality level.

[0033] By adopting the above technical solution, calculating the signal attenuation coefficient and adjusting the transmission power of the detection equipment accordingly, it is possible to reduce power waste while ensuring the quality of the detection signal and improve the detection energy efficiency ratio. Based on the atmospheric visibility data, the signal transmission quality level is determined and the operating frequency of the detection equipment is adjusted so that the detection signal frequency matches the current atmospheric transmission characteristics and reduces the attenuation and distortion of the signal during transmission. The dynamic adjustment mechanism of the transmission power and operating frequency of the detection equipment can keep the detection signal quality within a reasonable range at all times, improve the reliability of the detection data, improve energy utilization efficiency, and extend the service life of the detection equipment.

[0034] Furthermore, after controlling the drone to perform inspection along the inspection path, the method further includes:

[0035] Obtain the inspection data collected by the drone during the inspection process, and segment the inspection data according to the time dimension and space dimension to obtain segmented data;

[0036] Correlate and analyze the segmented data with the historical inspection data of each detection area to generate a dynamic parameter file of the inspection object;

[0037] Analyze the data distribution characteristics in the dynamic parameter files and identify the blind areas of inspection data collection based on the data distribution characteristics;

[0038] The location information of the blind spot is superimposed and analyzed with the three-dimensional grid inspection path to generate an optimized inspection path for supplementary collection.

[0039] By adopting the above technical solutions, the inspection data is segmented according to the time dimension and space dimension, and the dynamic parameter archive is generated by combining the historical inspection data for correlation analysis. The technical solution of identifying the collection blind area and generating the optimized inspection path by analyzing the data distribution characteristics makes the inspection data continuous and complete in the time and space dimensions. Through the processing and analysis of segmented data, the system can accurately locate the weak links and blind spots of data collection, and supplement the collection of these areas by optimizing the inspection path, thereby improving the coverage of inspection data. Through the establishment of dynamic parameter archives, the accurate characterization of the state of the inspection object is realized, which provides data support for the continuous improvement of inspection quality and improves the pertinence and effectiveness of inspection work.

[0040] Furthermore, the segmented data is correlated with the historical inspection data of each detection area to generate a dynamic parameter file of the inspection object, including:

[0041] Perform time series arrangement according to the timestamp information and spatial position information in the segmented data to obtain time series data;

[0042] Mapping the time series data to the corresponding detection area based on the spatial location information, and searching for the historical inspection data corresponding to the detection area;

[0043] Calculate the correlation index and data fluctuation characteristics between time series data and historical inspection data to obtain the change trend of inspection parameters;

[0044] Analyze the distribution patterns of inspection parameters at different time and space scales;

[0045] The inspection parameters are classified and labeled according to the distribution rules to obtain the dynamic parameter file of the inspection object.

[0046] By adopting the above technical solution, the segmented data is arranged in time series based on timestamp information and spatial location information and mapped to the corresponding detection area. The parameter change trend is obtained by calculating the correlation index and fluctuation characteristics with historical data, and the distribution law under different time and space scales is analyzed and classified and labeled. The technical solution makes the state characteristics of the inspection object more accurately characterized. By classifying and labeling the inspection parameters, the convenience of data retrieval and analysis is improved, the accurate evaluation of the state of the inspection object is supported, and the use value and analysis efficiency of the inspection data are improved.

[0047] In the second aspect, an embodiment of the present application provides a drone multi-mode mine inspection system, which drone multi-mode mine inspection system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call computer instructions to enable the system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0048] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a system, causes the system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0049] In a fourth aspect, an embodiment of the present application provides a computer program product, characterized in that when the computer program product runs on a system, the system executes the method described in any possible implementation manner in the first aspect.

[0050] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0051] 1. This application provides a multi-mode mine inspection method for drones, which divides the inspection targets into short-range objects and long-range objects. Combined with the setting of three-dimensional grid inspection paths and spare inspection entry points, differentiated inspection strategies can be formulated for inspection targets with different detection distances. By obtaining real-time environmental parameters such as light intensity, atmospheric visibility, electromagnetic field intensity, wind speed and direction, calculating the accessibility score of spare inspection entry points and selecting the optimal entry point, the drone can select the most suitable inspection path according to actual environmental conditions, improving the reliability and efficiency of inspection tasks, reducing the adverse effects of environmental factors on drone inspections, and at the same time enhancing the flexibility of the inspection path through the setting of spare entry points, so that the inspection system can better cope with complex and changeable mine environments, improving the integrity of inspection data, and improving the accuracy of safety assessment data obtained based on inspection data, thereby improving the safety of mines.

[0052] 2. This application provides a multi-mode mine inspection method using unmanned aerial vehicles, establishes a dynamic adjustment mechanism for detection equipment parameters and a real-time planning method for emergency evacuation routes, and enhances the environmental adaptability of the unmanned aerial vehicle inspection system. The system can dynamically compensate for the operating frequency and transmission power of the detection equipment according to environmental parameters, and by real-time monitoring of the changing trends of environmental parameters, timely calculate multiple candidate evacuation routes when environmental conditions deteriorate. By comprehensively considering the time cost and safety factor, the optimal evacuation route is selected, which not only improves the timeliness of the evacuation process, but also improves the safety of the evacuation process, thereby improving the inspection system's ability to respond to sudden environmental changes and reducing the risk of equipment damage and mission failure.

[0053] 3. This application provides a multi-mode mine inspection method using drones, which processes the inspection data in segments according to the time and space dimensions, generates dynamic parameter archives by combining historical inspection data for correlation analysis, and identifies collection blind spots by analyzing data distribution characteristics and generates a technical solution for optimizing inspection paths, so that the inspection data has continuity and integrity in the time and space dimensions. By processing and analyzing segmented data, the system can accurately locate weak links and blind spots in data collection, and supplement these areas by optimizing inspection paths, thereby improving the coverage of inspection data. By establishing dynamic parameter archives, the accurate characterization of the state of the inspection object is achieved, providing data support for the continuous improvement of inspection quality, and improving the pertinence and effectiveness of inspection work. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of a multi-mode mine inspection method using a drone in an embodiment of the present application.

[0055] Figure 2It is a flow chart of a detection equipment parameter adjustment and emergency evacuation method based on environmental parameters in an embodiment of the present application.

[0056] Figure 3 It is a schematic diagram of the physical device structure of a drone multi-mode mine inspection system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more listed items.

[0058] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0059] The following uses an embodiment and combines Figure 1 , a multi-mode mine inspection method using a drone in an embodiment of the present application is described:

[0060] See also Figure 1 , which is a flow chart of a multi-mode mine inspection method using a drone in an embodiment of the present application.

[0061] S101, dividing the inspection targets into short-range objects and long-range objects based on a preset inspection target classification list;

[0062] The system divides the inspection targets into short-range objects and long-range objects based on the preset inspection target classification list. Short-range objects include inspection targets whose detection distance is less than the preset distance, and long-range objects include inspection targets that need to be observed as a whole.

[0063] This step is to classify the target objects for drone inspection. The system pre-sets a list of inspection target classifications, which defines the characteristic attributes of various types of inspection objects. According to this classification list, the system divides the inspection targets into two categories: short-range objects and long-range objects. Among them, short-range objects refer to those inspection targets whose detection distance from the drone is less than the preset threshold, and usually require drones to perform close-range and refined detection; long-range objects refer to those large-scale inspection targets that require overall observation, and drones can perform global inspections on them at a longer distance.

[0064] In specific implementation, the system can use a rule-based classification method or a machine learning classification algorithm to achieve automatic classification of inspection targets. The rule-based classification method means that the system pre-sets a series of classification rules, then matches the attribute characteristics of each inspection target with the rules, and determines its category based on the matching results; the machine learning classification algorithm establishes a classification model through training sample data, and then inputs the inspection target into the model for predictive classification. The system can also introduce a manual intervention mechanism to allow users to adjust and correct the results of automatic classification to improve the accuracy and reliability of classification.

[0065] A technical problem that may arise during the inspection target classification process is that the attribute characteristics of some targets are unclear or incomplete, making it difficult to determine their category. To address this problem, the system can adopt the idea of ​​fuzzy classification, that is, allowing a target to belong to multiple categories at the same time, and assigning a membership value to each category to characterize its degree of belonging. The system can also handle unclassifiable targets by introducing default categories, classifying those targets with unclear attribute characteristics into a specific default class to ensure that each target can be properly classified.

[0066] S102, constructing a three-dimensional grid inspection path according to the location information of the near-range object and the far-range object, and setting a number of spare inspection entry points in the detection area of ​​the near-range object;

[0067] After the inspection target classification is completed, the system automatically generates a three-dimensional grid inspection path based on the spatial position distribution of short-range objects and long-range objects. This path needs to cover all areas to be inspected and take into account the efficiency and comprehensiveness of the inspection. At the same time, since short-range objects require close-range and refined inspection, the system will set up several spare inspection entry points in the inspection area of ​​each short-range object as alternative starting points for drones to enter the area for close-range inspection.

[0068] In specific implementation, the system can use a variety of path planning algorithms to construct a three-dimensional grid inspection path, such as the grid method, potential field method, sampling method, etc. Among them, the grid method is to divide the entire inspection space into a series of regular three-dimensional grids, and then search for an optimal inspection path on the grid; the potential field method is to regard each object to be inspected as an attractive potential field, and the drone flies in the direction of the minimum potential energy under the action of the potential field, thus naturally forming an inspection path; the sampling method is to randomly or according to certain rules in the inspection space. Sample a series of flight path points, and then find a shortest path connecting all sampling points through an optimization algorithm. For the setting of backup inspection entry points for short-range objects, the system can use a clustering algorithm to cluster the targets to be inspected in the detection area, and then set a backup entry point at the center of each cluster cluster to achieve full coverage detection of targets in the area.

[0069] S103, obtaining real-time environmental parameters;

[0070] The system obtains real-time environmental parameters, including light intensity, atmospheric visibility, electromagnetic field intensity, and wind speed and direction.

[0071] The system can obtain real-time environmental parameters through a variety of sensors and data sources. For example, the drone can be equipped with light intensity sensors, visibility sensors, electromagnetic field sensors, etc., and these sensors can be used to collect environmental data in real time; the environmental monitoring data of the current area can also be obtained from the meteorological department or other third-party agencies; the drone's own avionics equipment and flight control system can also be used to measure and calculate flight environmental parameters such as wind speed and direction.

[0072] S104, calculating the accessibility score of each backup inspection entrance point based on the degree of influence of the real-time environmental parameters on the drone, and determining the backup inspection entrance point with the highest accessibility score as the current inspection entrance point;

[0073] The system calculates the accessibility score of each backup inspection entry point based on the impact of real-time environmental parameters on the drone. Specifically, the real-time environmental parameters are input into the impact score calculation function to obtain the impact score; a three-dimensional vector is constructed based on the distance from the current position of the drone to each backup inspection entry point and the terrain height difference, and the flight energy consumption coefficient is calculated in combination with the impact score; the remaining power to reach each backup inspection entry point is calculated based on the flight energy consumption coefficient; the navigation risk value is calculated based on the impact score and the remaining power; the accessibility of each backup inspection entry point is evaluated based on the navigation risk value to obtain the accessibility benchmark score; the accessibility benchmark score and the historical success rate of each backup inspection entry point are weighted to obtain the accessibility score. Then, the system determines the backup inspection entry point with the highest accessibility score as the current inspection entry point.

[0074] Among them, the impact degree calculation function is: In the function, To measure the degree of impact, is the total duration of the inspection task, is the time variable, is the reference value of light intensity, for The light intensity at the time, is the reference value of visibility, for Visibility at all times, is the standard deviation of visibility, for The electromagnetic field strength at the time, is the influence function of electromagnetic field intensity, for The wind speed at the moment, for The wind direction angle at the moment, is the joint influence function of wind speed and wind direction, is the weight coefficient of light intensity, is the adjustment coefficient within the logarithmic function of light intensity, is the weight coefficient of visibility, is the weight coefficient of electromagnetic field strength, is the weight coefficient of wind speed and direction.

[0075] Represents the effect of light intensity on the drone. for The actual light intensity at the moment, is the reference value of light intensity. Calculate the degree of deviation of the light intensity from the reference value, add 1 and use it as The input is adjusted through the log function and then multiplied by the weight Get the impact score of light intensity.

[0076] Represents the impact of visibility on the UAV. for Actual visibility at the moment, It is the reference value of visibility. Calculate the standardized deviation of visibility from the reference value, square it and negate it, adjust the value range through the exponential function, and then multiply it by the weight Get the visibility impact score.

[0077] Indicates the impact of electromagnetic field strength on drones. for The actual electromagnetic field strength at the moment, is the unknown electromagnetic field strength influence function, multiplied by the weight ε to obtain the electromagnetic influence score.

[0078]

[0079] in, is the measured value of the electromagnetic field strength at time t, is the impact threshold of electromagnetic field strength, It is the steepness of the impact curve. When the electromagnetic field strength is lower than the threshold, the impact is small; when it is higher than the threshold, the impact increases rapidly and tends to 1. The value determines how quickly the curve goes from a small effect to a large effect.

[0080] Represents the combined impact of wind speed and direction on the UAV. and They are The wind speed and wind direction at the moment, is the combined influence function of wind speed and direction, multiplied by the weight Get the wind field impact score. Among them:

[0081]

[0082] in, for The measured wind speed at the time, for The measured value of wind direction angle at the time, is the flight direction angle of the drone, is the minimum threshold of wind speed influence, is the maximum threshold of wind speed influence, It is an indicator function, which takes the value 1 when the condition in the brackets is met, otherwise it takes the value 0.

[0083] In the function , , , , They are all weight coefficients, which are used to balance the weight of the impact of each environmental factor in the comprehensive impact score. is the standard deviation of visibility, which is used to standardize the deviation of visibility.

[0084] The sum of the above four items and the integration of time t from 0 to T give a comprehensive impact score S that reflects the environmental impact on the drone during the entire inspection process. The larger the S value, the greater the adverse impact of real-time environmental parameters on drone flight.

[0085] In this step, after obtaining real-time environmental parameters, the system evaluates the impact of these parameters on drone flight and inspection, and dynamically adjusts the inspection path and strategy to adapt to environmental changes. Specifically, the system calculates the flight risk and energy cost of the drone reaching each alternative inspection entry point based on environmental parameters, obtains an accessibility score, and then selects the entry point with the highest score (i.e., the lowest risk and the best accessibility) as the starting point of the current inspection.

[0086] In the process of calculating the impact score, the system uses a complex mathematical model to comprehensively consider the impact of multiple environmental factors such as light intensity, visibility, electromagnetic field intensity, wind speed and direction on the drone, and obtains a quantitative impact score through weighted summation. In the calculation of the accessibility score, the system not only considers the dynamic factor of the impact score, but also considers static factors such as the flight distance and altitude difference of the drone to the backup entry point, as well as the historical reachability of the entry point, and finally obtains a comprehensive accessibility score.

[0087] S105, generating an inspection path according to the current inspection entry point, and controlling the drone to perform inspection along the inspection path;

[0088] After determining the current inspection entry point, the system needs to generate a detailed inspection path based on the entry point and control the drone to perform the inspection task along the path. Generating an inspection path requires comprehensive consideration of multiple factors such as mission requirements, flight safety, and mission efficiency, and planning an optimal flight trajectory covering all target points based on the spatial distribution and priority of the inspection targets.

[0089] In specific implementation, the system can use a variety of path planning methods such as heuristic search, dynamic programming, and graph theory algorithms. Heuristic search usually adopts a greedy strategy, selecting the local optimal flight direction and target point each time, and constantly approaching the global optimal solution; dynamic programming decomposes the entire path planning problem into a series of sub-problems, and finds the optimal flight path through recursive solution and state transfer; graph theory algorithms abstract all target points into a weighted directed graph, and use classic algorithms such as minimum spanning tree and shortest path to generate an optimal path that traverses all nodes. After the system generates the inspection path, it needs to be further refined into a series of flight control instructions and sent to the UAV's flight control system to control the UAV's speed, altitude, heading and other flight parameters so that it flies strictly according to the planned path.

[0090] However, dynamic changes in the environment and emergencies can cause the pre-planned path to become invalid or no longer optimal. To address this problem, the system can introduce real-time path replanning and autonomous obstacle avoidance mechanisms. During the inspection process, the system obtains the flight status and environmental information of the drone in real time. When it deviates from the planned path or encounters an obstacle, it starts the path replanning program and dynamically generates a new local optimal path based on the current status and remaining target points to guide the drone to bypass the obstacle and return to the normal route. At the same time, the drone should also have the ability to avoid obstacles autonomously, sense obstacles ahead through onboard sensors, and automatically plan obstacle avoidance paths to ensure flight safety.

[0091] S106, obtaining inspection data collected by the drone during the inspection process, and segmenting the inspection data according to the time dimension and the space dimension to obtain segmented data;

[0092] When the drone is inspecting along the planned route, the system obtains various inspection data collected by the drone in real time, such as images, videos, environmental parameters, etc. In order to facilitate subsequent data analysis and processing, the system needs to segment the original inspection data and divide the data into independent data segments according to the two dimensions of time and space.

[0093] In specific implementation, the system can use methods such as sliding windows and recursive partitioning to segment time series data. Sliding windows refer to setting a time window of fixed length, sliding the window on the time axis, and extracting the data in the window to form a data segment; recursive segmentation uses a threshold or split point to recursively divide the time series data into two left and right subsequences until the length or characteristics of each subsequence meet the requirements. In the spatial dimension, the system can map the inspection data to geographic grids or regional blocks based on the GPS trajectory information of the drone, and each grid or region corresponds to a spatial data segment. When dividing the data segments, the system needs to fully consider the continuity, integrity and interpretability of the data, to ensure the independence of each data segment, and to avoid information loss and semantic confusion caused by excessive cutting.

[0094] S107, performing correlation analysis on the segmented data and the historical inspection data of each detection area to generate a dynamic parameter file of the inspection object;

[0095] The system correlates the segmented data with the historical inspection data of each detection area and generates a dynamic parameter archive of the inspection object. Specifically: the time series data is arranged according to the timestamp information and spatial location information in the segmented data to obtain the time series data; the time series data is mapped to the corresponding detection area based on the spatial location information, and the historical inspection data corresponding to the detection area is searched; the correlation index and data fluctuation characteristics between the time series data and the historical inspection data are calculated to obtain the changing trend of the inspection parameters; the distribution law of the inspection parameters at different time and space scales is analyzed; the inspection parameters are classified and labeled according to the distribution law to obtain the dynamic parameter archive of the inspection object.

[0096] After obtaining the segmented results of the inspection data, the system needs to compare and analyze each segmented data with the historical inspection data of the corresponding detection area, examine the dynamic changes of the target parameters, and form a complete dynamic parameter file of the inspection object. By tracking and analyzing the changes in the object parameters over time and space, the dynamic characteristics of the inspection object can be mastered, providing a basis for the detection and early warning of abnormal situations.

[0097] In specific implementation, the system can use time series analysis, spatial statistical analysis and other methods to analyze the correlation between each segmented data and historical data from the perspectives of time and space. In the time dimension, the system can extract the time series characteristics of each parameter, such as trends, cycles, mutations, etc., and find out the corresponding relationship between current data and historical data in time through similarity measurement and correlation analysis; in the spatial dimension, the system can use spatial interpolation, spatial clustering and other methods to analyze the distribution pattern and correlation structure of parameters in space, and examine the correlation between current data and historical data in spatial position. On this basis, the system can combine the analysis results of the two dimensions of time and space, calculate the comprehensive correlation strength between each segmented data and historical data, and generate a dynamic parameter change archive.

[0098] S108, analyzing the data distribution characteristics in the dynamic parameter file, and identifying the blind area for collecting inspection data according to the data distribution characteristics;

[0099] After generating the dynamic parameter archive of the inspection object, the system needs to further analyze the distribution characteristics of the archive data, examine the data collection density and coverage in time and space, and identify possible collection blind spots. The so-called collection blind spot refers to the lack of data collection or lack of data in certain time periods or spatial areas due to various reasons, forming a blank spot in monitoring. Finding the collection blind spot can help the system optimize the sensor layout and inspection route planning, and improve the comprehensiveness and effectiveness of monitoring.

[0100] In specific implementation, the system can use methods such as statistical analysis and machine learning to analyze the distribution characteristics of data from multiple perspectives. For example, the system can count the amount of data and density distribution in each time period and spatial grid, and find areas with fewer sampling points or significantly lower density; it can also calculate the statistical characteristics of the data, such as mean, variance, skewness, etc., to examine the concentration and dispersion of data distribution and find abnormal areas; in addition, the system can also use machine learning algorithms such as clustering and anomaly detection to automatically divide different data clusters according to the similarity and difference of the data, and identify isolated points or abnormal areas that are significantly different from most patterns. These areas with abnormal distribution or statistical abnormalities often correspond to monitoring blind spots and require the system to focus on them.

[0101] However, the data distribution of different regions and targets varies greatly, and the use of a unified anomaly discrimination standard will reduce the accuracy of recognition. To address this problem, the system can introduce adaptive thresholds and hierarchical detection mechanisms. The system can automatically learn and adjust the anomaly judgment threshold for different regions and targets, taking into full account the dynamic changes in data distribution and regional differences; at the same time, the system can adopt a hierarchical detection strategy, first roughly identifying the abnormal area as a whole, and then finely locating and verifying the anomaly locally, to improve the accuracy and efficiency of detection. In addition, the system can also make full use of multi-source heterogeneous data, and reduce the risk of misjudgment of single data through complementary verification and cross-analysis.

[0102] S109: Superimpose and analyze the location information of the acquisition blind area with the three-dimensional grid inspection path to generate an optimized inspection path for supplementary acquisition.

[0103] After identifying the blind spots for collecting inspection data, the system needs to plan a supplementary inspection route for these blind spots and focus on monitoring the targets in the blind spots to make up for the shortcomings of the original route. The system first overlays and analyzes the spatial location information of the blind spots with the original three-dimensional grid inspection route to find out the spatial relationship between the route and the blind spots and examine whether the blind spots can be effectively covered by the original route. For blind spots that are not covered or are insufficiently covered, the system will plan a special supplementary inspection route to ensure focused monitoring of the area.

[0104] In specific implementation, the system can use multi-objective optimization, heuristic search and other methods to generate an optimal path that takes into account both blind spot coverage and inspection efficiency. Among them, multi-objective optimization can use multiple factors such as blind spot coverage, flight distance, energy consumption, etc. as optimization targets, and find a set of balanced and effective alternative paths through weighted summation and other methods; heuristic search algorithms can flexibly explore in the path space, using heuristic rules such as greedy strategies and simulated annealing to quickly search for the optimal solution. At the same time, the system also needs to consider realistic factors such as the flight dynamics constraints of the drone and the distribution of environmental obstacles when planning the path, and optimize the path in terms of feasibility and safety.

[0105] In the above embodiment, the inspection targets are divided into short-range objects and long-range objects. Combined with the setting of three-dimensional grid inspection paths and spare inspection entry points, differentiated inspection strategies can be formulated for inspection targets with different detection distances. By obtaining real-time environmental parameters such as light intensity, atmospheric visibility, electromagnetic field intensity, wind speed and direction, calculating the accessibility score of spare inspection entry points and selecting the optimal entry point, the drone can select the most suitable inspection path according to the actual environmental conditions, improving the reliability and efficiency of the inspection task, reducing the adverse effects of environmental factors on drone inspections, and at the same time enhancing the flexibility of the inspection path through the setting of spare entry points, so that the inspection system can better cope with the complex and changeable mining environment, improving the integrity of the inspection data, and improving the accuracy of the safety assessment data obtained based on the inspection data, thereby improving the safety of the mine.

[0106] Further, in another embodiment, the mining geological model data and historical disaster data are obtained, and the mining area is divided into three-dimensional space based on the geological model data to obtain a number of three-dimensional inspection units; in this step, the system first needs to obtain the geological model data describing the spatial attributes of the mining area, such as the geological structure, topography and landform of the mining area, as well as the historical record data of disaster events such as collapse, landslide, and mudslide in the mining area. Then, based on the mining area geological model, the system comprehensively considers factors such as terrain undulation, rock layer distribution, and fault structure, and divides the entire mining area space into a series of regular or irregular three-dimensional areas. Each area is defined as an independent inspection unit as the basic unit for subsequent inspection task allocation and path planning. When dividing the inspection unit, the system considers the similarity of the geological attributes of each unit and the probability of disaster occurrence, and tries to keep the geological conditions and disaster risk levels within the same unit consistent, and the differences between different units are more obvious, laying the foundation for subsequent targeted inspections in different regions.

[0107] The three-dimensional inspection unit is spatially associated with the historical disaster data, the disaster risk index of each inspection unit is calculated, and the inspection priority list is generated in order from high to low according to the disaster risk index. After obtaining the three-dimensional inspection unit, the system matches and associates the spatial location information of each unit with the historical disaster data, and counts the number, frequency, intensity and other information of various disaster events that have occurred in each unit. On this basis, the system constructs a comprehensive evaluation model, taking into account the harmfulness of different types of disasters, the vulnerability of different regions and the influence weights of various inducing factors, and calculates the disaster risk index of each inspection unit. The index reflects the possibility of disasters in each unit. The higher the index, the higher the disaster risk in the area, which needs to be paid special attention during the inspection. According to the evaluation results of the disaster risk index, the system generates a priority ranking list of inspection units, with units with high indexes in front and units with low indexes in the back, providing a regional priority reference for drone inspections.

[0108] Obtain real-time monitoring data of the mine environment, and determine the optimal inspection unit sequence in combination with the inspection priority list. After generating the inspection priority list, the system also needs to obtain real-time monitoring data of the mine environment, including environmental parameter information collected by multi-source sensors such as meteorology, hydrology, and geology, to evaluate the current changes in disaster risks in various regions. According to the results of real-time environmental data analysis, the system dynamically adjusts the priority level of the inspection unit to match the current risk level. For example, when a certain area monitors abnormal conditions such as a sudden increase in rainfall and a rise in groundwater levels, the system determines that the risk of landslides and mudslides in the area is increasing, and automatically increases the ranking of the area in the priority list; conversely, if all monitoring indicators in a certain area are within the normal range and remain stable for a long time, the system appropriately reduces the priority level of the area. Through real-time environmental monitoring and dynamic priority adjustment, the system determines a dynamically updated optimal inspection unit sequence to guide drones to fly to high-risk areas for key monitoring.

[0109] According to the optimal inspection unit sequence and the spatial topological relationship of the three-dimensional inspection units, a three-dimensional inspection path is constructed with a greedy strategy. After determining the optimal inspection unit sequence, the system needs to further construct a continuous and efficient three-dimensional inspection path based on the topological relationship of each unit in three-dimensional space. Due to the large number of inspection units and the limited flight time and range of the drone, the path planning must take into account the inspection priority and flight efficiency. It is necessary to give priority to covering high-risk areas, and to shorten the flight distance as much as possible and avoid repeated scanning. The system adopts a greedy strategy, and each time selects the unit closest to the current position, with the highest priority and not inspected as the next flight target until all units are covered or the drone runs out of power. When generating the inspection path, the system makes full use of the spatial adjacency of the inspection units, tries to avoid the drone from making large turns or flying long distances across regions, and improves the continuity and accessibility of the path. At the same time, the system also considers the flight performance and airspace restrictions of the drone, and performs smooth optimization and safety verification on the route to ensure that the generated three-dimensional inspection path meets the requirements of the actual execution of the drone.

[0110] Control the drone to fly along the three-dimensional inspection path to obtain high-resolution remote sensing images of the inspection unit. The system sends the planned three-dimensional inspection path to the drone, and the autonomous flight control system of the drone parses the path and converts it into a series of waypoint coordinates and flight action instructions, controls the drone to fly according to the specified route and altitude, and scans and images each inspection unit in turn. During the flight, the high-resolution optical or multi-spectral camera carried by the drone collects detailed surface information of each inspection unit, and obtains high-definition image data of the surface morphology, distribution of objects, vegetation growth and other conditions of key areas such as open-pit mines, spoil dumps, and tailings ponds. In order to ensure the availability and complementarity of image data, the system adaptively adjusts the spatial resolution and overlap of the camera according to the size of the inspection unit and the complexity of the objects, avoiding data redundancy and omissions, and controlling the amount of data not to exceed the storage and transmission capacity. After the drone completes the flight shooting task of each unit, it transmits the image data back to the ground station in real time, which is automatically received and stored by the system.

[0111] The remote sensing images are geometrically corrected, enhanced and feature extracted, and a three-dimensional real-scene model of the inspection unit is constructed. Due to the changes in the flight attitude and imaging angle of the drone, the original remote sensing images obtained often have problems of geometric distortion and uneven grayscale, which makes it difficult to directly use them for target interpretation and change analysis. Therefore, the system first needs to geometrically correct the image data to eliminate distortion and restore the true shape and spatial relationship of the objects. The system can use high-precision GPS / INS combined navigation and positioning data to establish a mapping relationship from image coordinates to ground coordinates based on the real-time attitude and camera parameters of the drone to achieve accurate geometric correction. On this basis, the system uses image processing methods such as histogram equalization and edge enhancement to improve the visual effect of the image and highlight the texture and contour features of the target of interest. Then, the system uses computer vision and image analysis algorithms to extract key features in the image, such as color, texture, shape, etc., and segment, identify and reconstruct the ground objects based on this. Finally, the system integrates the corrected and enhanced images of each inspection unit with its three-dimensional spatial coordinates to construct a realistic three-dimensional real-scene model, intuitively presenting the three-dimensional morphology and spatial relationship of the mining area's surface environment, and providing a reliable image interpretation basis for disaster risk investigation and trend warning.

[0112] Combined with the mine geological model, the three-dimensional real-life model is analyzed in time and space to identify the early signs of disaster hazards such as landslides, collapses, and mud-rock flows, and to build a disaster warning indicator system. Based on the three-dimensional real-life model of the inspection unit, the system is further combined with background models such as the geological structure and hydrogeological conditions of the mining area to conduct a comprehensive analysis from the two dimensions of time and space. In the time dimension, the system analyzes the dynamic changes of factors such as surface morphology, distribution of land objects, and vegetation growth by comparing images of different periods, and identifies the early signs of disaster hazards such as landslides, collapses, and mud-rock flows, such as mountain cracks, building deformation, and vegetation degradation. In the spatial dimension, the system combines spatial information such as geological structure, rock layer distribution, and fault direction to analyze the spatial differentiation laws and evolution trends of disaster hazards and delineate high-risk areas. Based on the rich information on signs of disaster hazards, the system constructs a multi-factor, multi-scale disaster warning indicator system based on the mechanism of disasters, covering multiple warning factors such as surface deformation, vegetation anomalies, and hydrological anomalies. Then, the system sets warning thresholds for each type of indicator, establishes a mapping relationship from signs to warning levels, and forms a scientific basis for quantitatively evaluating the possibility and danger of disasters.

[0113] Comprehensively analyze the spatiotemporal dynamic changes of various early warning indicators, and make accurate early warnings for the time, location and degree of harm of mine disasters. After establishing the disaster early warning indicator system, the system needs to dynamically track and analyze the spatiotemporal distribution laws and evolution trends of various indicators. The system obtains the monitoring data of the multi-source early warning indicators of each inspection unit in real time, and describes the characteristics of the change of indicators in time and space through time series analysis, spatial correlation analysis and other methods. On the one hand, the system analyzes the dynamic characteristics of the change rate and acceleration of each indicator, captures the sudden change and abnormality of the indicator during the disaster breeding process, and estimates the time window for the occurrence of the disaster; on the other hand, the system examines the spatial correlation of early warning indicators in different regions, identifies the impact range and potential destructive power of the disaster, and delineates the dangerous area. Based on the comprehensive analysis, the system automatically triggers the corresponding level of early warning information according to the degree of danger and response mechanism of the early warning indicators, and intuitively marks the spatial location and range of the early warning area through a three-dimensional real-life model. Finally, the system forms a set of accurate mine disaster early warning solutions, predicts the time, location and harm intensity of the disaster in advance, and provides scientific support for disaster prevention and control decision-making and emergency response.

[0114] In the above embodiments, by introducing a three-dimensional inspection unit and a disaster risk assessment mechanism, dynamic monitoring of the mining environment in different regions and key areas is achieved, thereby improving the pertinence and efficiency of drone inspections; by constructing a three-dimensional real-life model through high-resolution remote sensing images, and combining spatiotemporal data mining to analyze multi-source early warning information of disaster hazards, the adequacy and reliability of disaster early warnings are enhanced; by comprehensively analyzing multi-factor early warning indicators, the spatiotemporal distribution and degree of harm of disasters are quantitatively characterized and accurately predicted, thereby enhancing the timeliness and accuracy of disaster early warnings. Therefore, the technical solution of this embodiment has better solved the problems of the lack of pertinence, insufficient data analysis, and inaccurate early warning of existing mine drone inspections, and achieved the goals of refined monitoring, early warning, and dynamic prevention and control of mine disasters.

[0115] In the first embodiment, the drone needs to dynamically adjust the inspection path according to the actual environmental conditions when performing the inspection task. In order to further improve the inspection effect and respond to sudden environmental changes, the present application also provides a detection equipment parameter adjustment and emergency evacuation method based on environmental parameters. This method adjusts the working parameters of the detection equipment in real time to adapt to environmental changes, and plans the optimal evacuation path in time when the environmental conditions deteriorate, thereby ensuring the safety and reliability of the inspection task. Figure 2 , a detection equipment parameter adjustment and emergency evacuation method based on environmental parameters in an embodiment of the present application is described:

[0116] See also Figure 2 , which is a flow chart of a detection equipment parameter adjustment and emergency evacuation method based on environmental parameters in an embodiment of the present application.

[0117] S201, calculating the signal attenuation coefficient according to the electromagnetic field strength data in the real-time environmental parameters;

[0118] This step is to analyze and process the electromagnetic field strength data obtained in real time, and calculate the degree of signal attenuation in the current environment. Electromagnetic field strength is one of the important factors affecting the quality of radio signal transmission. By real-time monitoring of electromagnetic field strength, the propagation characteristics of electromagnetic waves in the current environment can be mastered. The system receives electromagnetic field strength data from the environmental perception module, and by establishing a mathematical model of electromagnetic field strength and signal attenuation, a coefficient that quantitatively calculates the degree of signal attenuation is calculated.

[0119] The specific implementation method of this step can be: the system constructs a mapping table of electromagnetic field strength and signal attenuation coefficient, and quickly obtains the corresponding attenuation coefficient by looking up the table or interpolation calculation according to the measured electromagnetic field strength value; or, the system summarizes a functional relationship between electromagnetic field strength and signal attenuation coefficient through a large number of experiments and theoretical analysis in advance, and directly substitutes the formula for calculation every time new electromagnetic field strength data is received. In addition, the system can also select a targeted calculation model according to the specific environment where the drone is located (such as urban areas, suburbs, mountains, etc.) to further improve the accuracy of attenuation coefficient calculation.

[0120] S202, adjusting the transmission power of the detection device based on the signal attenuation coefficient;

[0121] This step uses the signal attenuation coefficient calculated in the previous step to dynamically adjust the transmission power of the drone detection equipment to compensate for the signal attenuation caused by the environment and ensure communication quality. Transmission power is a key working parameter of the detection equipment, which determines the radiation intensity and effective coverage range of electromagnetic waves. When the signal attenuation increases, appropriately increasing the transmission power can improve the signal's penetration and anti-interference ability; conversely, when the signal attenuation decreases, reducing the transmission power can both save energy and reduce electromagnetic pollution to the environment.

[0122] The system can realize adaptive adjustment of the transmission power of the detection equipment by adopting the following specific technical solutions: First, the system determines the theoretical adjustment range of the transmission power according to the signal attenuation coefficient and divides it into several gears; then, the system monitors the working status and environmental parameters of the detection equipment in real time, comprehensively evaluates the current communication quality, and selects the optimal power gear; finally, the new power parameters are sent to the detection equipment through the hardware or software interface to guide it to make corresponding power adjustments. During the adjustment process, the system also needs to consider the physical characteristics and safety limitations of the equipment itself to avoid problems such as overload or damage.

[0123] S203, determining the signal transmission quality level according to the atmospheric visibility data in the real-time environmental parameters;

[0124] This step uses the visibility data obtained in real time to evaluate the impact of the current atmospheric environment on radio wave propagation, and then judge the quality of signal transmission. Visibility is an important indicator of air transparency, which is closely related to factors such as suspended particles and water vapor content in the atmosphere. Under conditions of low visibility, electromagnetic waves are easily affected by effects such as absorption and scattering, which leads to problems such as signal strength attenuation and phase distortion, and the quality of transmission is reduced.

[0125] To achieve this step, the system first needs to establish a corresponding relationship model between visibility and signal transmission quality. Through a large number of field tests and statistical analysis, the typical characteristics of radio wave propagation under different visibility conditions can be summarized, and a set of signal transmission quality grading standards can be formulated accordingly, such as excellent, good, medium, poor, etc. In actual applications, the system obtains visibility data fed back by the environmental perception module in real time, and quickly determines the current signal transmission quality level by querying the established corresponding relationship model.

[0126] S204, adjusting the operating frequency of the detection device based on the signal transmission quality level;

[0127] This step is to dynamically select the optimal operating frequency of the detection equipment based on the signal transmission quality level obtained in the previous step to adapt to the current transmission environment and improve the reliability of data transmission. In the process of electromagnetic wave propagation, signals of different frequencies have different physical properties, such as penetration, diffraction ability, anti-interference, etc. By specifically adjusting the operating frequency of the detection equipment, the impact of harsh environment on signal transmission can be avoided or reduced to a certain extent.

[0128] To implement this step, the system needs to pre-plan the optional operating frequency set of the detection equipment and formulate corresponding applicable conditions and pros and cons evaluation indicators for each frequency. When a new signal transmission quality level is obtained, the system traverses all alternative frequencies in the current environment according to the established frequency selection strategy, and finds the best operating frequency by comprehensively evaluating factors such as signal transmission performance, anti-interference ability, and energy consumption level, and configures it to the detection equipment. During the frequency adjustment process, the system also needs to consider constraints such as the frequency adaptation range of the equipment hardware and the occupancy of channel resources to ensure the feasibility and compliance of the adjustment plan.

[0129] S205, acquiring sampling data of real-time environmental parameters in real time, and calculating the change amount of the environmental parameters between two adjacent sampling moments;

[0130] This step is to monitor and analyze the changes in the environment in which the drone is located in real time. By continuously collecting environmental parameter data and calculating the parameter changes between adjacent sampling moments, abnormal fluctuations in the environment can be discovered in a timely manner. Environmental parameters usually include temperature, humidity, air pressure, wind speed and other physical quantities. Changes in these parameters will have a direct or indirect impact on the flight and detection missions of the drone.

[0131] To achieve this step, the system needs to be equipped with a series of high-precision, fast-response environmental sensors to continuously sample and record the environmental parameters around the drone. The selection of sampling frequency needs to comprehensively consider factors such as the time scale of parameter changes, sensor performance, and data processing capabilities. It must not only meet the sensitivity requirements of anomaly detection, but also take into account the efficiency and stability of the system. After obtaining continuous sampling data, the system compares the data of two adjacent sampling moments item by item, calculates the change in each parameter, and obtains a quantitative indicator reflecting the rate and magnitude of environmental changes.

[0132] In order to accurately judge whether there are abnormal changes in the environment, the system also needs to set a reasonable threshold for the change amount. The fluctuation range of parameters within the normal range can be determined in advance according to the design indicators of the drone, the type of flight mission, historical experience, etc. When the change amount exceeds this range, it is considered that an abnormal situation has occurred. At the same time, the system can also automatically optimize the setting of thresholds through statistical analysis and machine learning technologies. With the continuous accumulation of data and the dynamic changes of the environment, the abnormal judgment conditions can be adaptively adjusted to improve the intelligence level of the system. In addition, when calculating the change amount, the system also needs to consider the credibility and continuity of the data, identify and process abnormal situations such as sensor failure and signal loss, and ensure the reliability of the change amount calculation. The system monitors the changes in environmental parameters in real time, automatically optimizes the abnormal judgment threshold, and ensures the reliability of data, which can enhance the perception and adaptability of drones to complex environments and make emergency disposal in time.

[0133] S206, when the change in the environmental parameter exceeds a preset threshold, calculating a number of candidate evacuation paths based on the backup inspection entry point;

[0134] This step is to trigger the emergency evacuation mechanism of the drone when a drastic change in the environment is detected, and immediately start the calculation and evaluation of the evacuation path. When the change in environmental parameters exceeds the pre-set warning threshold, it means that the current environmental conditions are no longer suitable for continuing the established inspection mission, and the drone needs to be transferred to a safe area as soon as possible to avoid potential dangers.

[0135] To achieve this step, the system first needs to maintain a database of alternative inspection entry points. These entry points are usually selected in areas with relatively stable environments and good conditions, which can provide temporary shelter and supplies for drones. When receiving the evacuation command, the system immediately loads the location information of all candidate entry points, and combines the current location and flight status of the drone to calculate the flight path from the current location to each entry point. The generation of the path needs to comprehensively consider factors such as flight distance, terrain undulations, and obstacle distribution, and choose a safe and efficient route as much as possible.

[0136] Since environmental changes may affect multiple alternative entry points, the system usually calculates several different candidate paths as alternatives when generating an evacuation path. These candidate paths may differ in terms of evacuation target points, route directions, etc., providing more options for subsequent path optimization. In the process of calculating candidate paths, the system also needs to update environmental parameters and drone status information in real time, dynamically adjust the feasibility assessment of the path, and ensure that the generated evacuation path always meets the requirements of safety and real-time performance.

[0137] In a complex environment, a single evacuation path may face greater uncertainty risks. In order to further improve the success rate of emergency evacuation, the system can also introduce intelligent game theory and risk assessment models to select an optimal path combination from multiple candidate paths to form a collective and diversified evacuation plan. At the same time, the system can also dynamically adjust the weight and priority of the evacuation plan according to real-time environmental changes, so that the path selection can adapt to new situations at any time. The system uses intelligent optimization technology to generate the optimal evacuation path combination and implements dynamic adjustments, which can significantly improve the safety and reliability of drone emergency evacuation.

[0138] S207: Calculate the time costs and safety factors of several candidate evacuation paths, and select the evacuation path with the lowest time cost and a safety factor that meets preset conditions as the optimal evacuation path.

[0139] This step is to select the optimal evacuation path after obtaining several candidate evacuation paths by comprehensively evaluating the time cost and safety factor of each path. Since the main contradiction faced by drones during emergency evacuation is the balance between "speed" and "safety", the system needs to find an optimal path that can reach the target point in the shortest possible time with the lowest risk along the way.

[0140] To achieve this step, the system needs to calculate the time cost and safety factor of each candidate path respectively. Among them, the time cost mainly refers to the time required for the drone to fly to the target point along the path, which is related to factors such as the spatial distance of the path, flight speed, and environmental resistance. The system can establish a flight dynamics model to simulate the flight trajectory and time consumption of the drone under different paths, and derive the theoretical time cost of each path. The safety factor is a dimensionless relative indicator used to measure the comprehensive safety risk level along the path. It needs to consider multiple factors such as terrain complexity, obstacle density, and environmental interference intensity, and quantify various security threats into a normalized safety factor value through weighted summation or fuzzy evaluation methods.

[0141] After obtaining the time cost and safety factor of the candidate paths, the system sorts and screens all paths according to the principle of "optimal time, safe and controllable". First, in order to ensure the rapid response of the evacuation task, the system will give priority to the path with the lowest time cost; then, on the basis of meeting the time requirements, the system further checks the safety factor of the path. If the safety factor is lower than the preset safety threshold, the path will be judged as unacceptable and automatically screened out. Finally, the system selects the candidate path with the best time cost and the safety factor that meets the standard as the optimal path for this evacuation task.

[0142] In the above embodiment, the inspection data is segmented according to the time dimension and the space dimension, and the dynamic parameter archive is generated by combining the historical inspection data for correlation analysis. The technical solution of identifying the collection blind area and generating the optimized inspection path by analyzing the data distribution characteristics makes the inspection data continuous and complete in the time and space dimensions. Through the processing and analysis of the segmented data, the system can accurately locate the weak links and blind spots of data collection, and supplement the collection of these areas by optimizing the inspection path, thereby improving the coverage of the inspection data. Through the establishment of the dynamic parameter archive, the accurate characterization of the state of the inspection object is realized, which provides data support for the continuous improvement of the inspection quality and improves the pertinence and effectiveness of the inspection work.

[0143] The following describes the system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of the physical device structure of a drone multi-mode mine inspection system provided in an embodiment of the present application.

[0144] It should be noted that Figure 3 The structure of the system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0145] like Figure 3As shown, the system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through a bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0146] The following components are connected to the I / O interface 305: an input section 306 including a camera, an infrared sensor, etc.; an output section 307 including a liquid crystal display (LCD) and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0147] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0148] It should be noted that the computer-readable medium shown in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.

[0149] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Among them, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0150] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the system described in the above embodiment; or may exist independently without being assembled into the system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of a system, the system implements the method provided in the above embodiment.

[0151] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0152] As used in the above embodiments, the term "when..." may be interpreted as "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted as "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0153] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0154] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A multi-mode mine inspection method using an unmanned aerial vehicle, characterized in that: include: Based on a preset inspection target classification list, the inspection targets are divided into short-range objects and long-range objects, wherein the short-range objects include inspection targets whose detection distance is less than a preset distance, and the long-range objects include inspection targets that need to be observed as a whole; Constructing a three-dimensional grid inspection path according to the position information of the short-range object and the long-range object, and setting a plurality of spare inspection entry points in the detection area of ​​the short-range object; Acquiring real-time environmental parameters, including light intensity, atmospheric visibility, electromagnetic field intensity, and wind speed and direction; The accessibility score of each of the backup inspection entry points is calculated based on the degree of influence of the real-time environmental parameters on the drone, and the backup inspection entry point with the highest accessibility score is determined as the current inspection entry point. The influence degree calculation function is: In the function, S is the impact degree score, T is the total duration of the inspection task, t is the time variable, and I ref is the reference value of the light intensity, I(t) is the light intensity at time t, and V ref is the reference value of visibility, V(t) is the visibility at time t, and σ v is the standard deviation of visibility, E(t) is the electromagnetic field intensity at the time t, f(E(t)) is the influence function of the electromagnetic field intensity, W(t) is the wind speed at the time t, φ(t) is the wind direction angle at the time t, g(W(t), φ(t)) is the joint influence function of wind speed and wind direction, α is the weight coefficient of the light intensity, β is the adjustment coefficient in the logarithmic function of the light intensity, γ is the weight coefficient of the visibility, ε is the weight coefficient of the electromagnetic field intensity, and η is the weight coefficient of the wind speed and wind direction influence term; An inspection path is generated according to the current inspection entry point, and the UAV is controlled to perform inspection along the inspection path.

2. The method according to claim 1, characterized in that The calculating of the accessibility score of each of the backup inspection entry points based on the degree of influence of the real-time environmental parameters on the drone specifically includes: Inputting the real-time environmental parameters into an impact degree score calculation function to obtain an impact degree score; Constructing a three-dimensional vector according to the distance from the current position of the UAV to each of the backup inspection entry points and the terrain height difference, and calculating the flight energy consumption coefficient in combination with the degree of influence; Calculate the remaining power reaching each of the standby inspection entry points based on the flight energy consumption coefficient; Calculate the navigation risk value according to the impact degree and the remaining power; Evaluate the accessibility of each of the alternative inspection entry points based on the navigation risk value to obtain an accessibility benchmark score; The accessibility benchmark score and the historical success rate of each of the standby inspection entry points are weighted and calculated to obtain an accessibility score.

3. The method according to claim 1, characterized in that After controlling the drone to perform inspection along the inspection path, the method further includes: Based on the real-time environmental parameters, the detection device of the UAV is adjusted in parameters, wherein the parameter adjustment includes adjusting the working frequency and the transmission power of the detection device to obtain dynamic compensation parameters; Acquire the sampling data of the real-time environmental parameters in real time, and calculate the change amount of the environmental parameters at two adjacent sampling moments; When the environmental parameter change exceeds a preset threshold, a plurality of candidate evacuation paths are calculated based on the backup inspection entry point; The time costs and safety factors of the plurality of candidate evacuation paths are calculated, and the evacuation path with the lowest time cost and the safety factor satisfying a preset condition is selected as the optimal evacuation path.

4. The method according to claim 3, characterized in that The step of adjusting the parameters of the detection equipment of the UAV based on the real-time environmental parameters specifically includes: Calculating a signal attenuation coefficient according to the electromagnetic field strength data in the real-time environmental parameters; adjusting the transmission power of the detection device based on the signal attenuation coefficient; Determining the signal transmission quality level according to the atmospheric visibility data in the real-time environmental parameters; The operating frequency of the detection device is adjusted based on the signal transmission quality level.

5. The method according to claim 1, characterized in that After controlling the drone to perform inspection along the inspection path, the method further includes: Acquire the inspection data collected by the drone during the inspection process, and segment the inspection data according to the time dimension and the space dimension to obtain segmented data; Correlate and analyze the segmented data with the historical inspection data of each detection area to generate a dynamic parameter file of the inspection object; Analyzing the data distribution characteristics in the dynamic parameter file, and identifying the blind area for collecting the inspection data according to the data distribution characteristics; The location information of the acquisition blind area is superimposed and analyzed with the three-dimensional grid inspection path to generate an optimized inspection path for supplementary acquisition.

6. The method according to claim 5, characterized in that The step of correlating and analyzing the segmented data with the historical inspection data of each detection area to generate a dynamic parameter file of the inspection object specifically includes: Performing time series arrangement according to the timestamp information and spatial position information in the segmented data to obtain time series data; Mapping the time series data to a corresponding detection area based on the spatial location information, and searching for historical inspection data corresponding to the detection area; Calculate the correlation index and data fluctuation characteristics between the time series data and the historical inspection data to obtain the change trend of the inspection parameters; Analyze the distribution patterns of the inspection parameters at different time and space scales; The inspection parameters are classified and labeled according to the distribution law to obtain a dynamic parameter file of the inspection object.

7. A multi-mode mine inspection system using unmanned aerial vehicles, characterized in that: The system comprises: One or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the system to execute the method as described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a system, the system is caused to execute the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product is executed on a system, the system is caused to execute the method according to any one of claims 1 to 6.

Images