An unmanned aerial vehicle inspection system for environmental condition monitoring

Abstract

This invention discloses a drone inspection system for environmental status monitoring, relating to the field of drone control technology. It includes a pollutant monitoring data acquisition module, a pollutant monitoring data analysis module, a pollution diffusion point determination module, a pollution diffusion trend line acquisition module, an inspection point coordinate acquisition module, and an inspection route acquisition module. By generating an optimal or suboptimal inspection route, the drone can efficiently pass through these predicted inspection points sequentially, achieving accurate tracking and monitoring of the predicted pollution diffusion path. This ensures that the drone remains at the forefront or key area of ​​pollution diffusion. Through the sorting of predicted points and path planning, an efficient inspection route can be generated, reducing unnecessary detours. This allows the drone to adapt to changes in the pollution diffusion path in real time, ensuring the effectiveness and timeliness of monitoring. In the event of a sudden pollution incident involving the drone, the inspection route can be quickly adjusted to focus on monitoring the pollution diffusion area.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) control technology, specifically a UAV inspection system for environmental condition monitoring. Background Technology

[0002] With rapid industrialization and urbanization, environmental pollution has become increasingly serious, making real-time and accurate monitoring of the environment crucial. Traditional environmental monitoring methods typically rely on the deployment of fixed monitoring stations. These stations have limited distribution and coverage, making it difficult to comprehensively cover the monitored area and track the spread of pollutants in real time. In recent years, drone technology has been widely adopted in environmental monitoring. Drones offer advantages such as high mobility, wide coverage, and flexible deployment, and can be equipped with various sensors to sample and monitor environmental pollutants.

[0003] However, most existing UAV environmental monitoring systems can only perform simple pollutant concentration measurements and cannot effectively predict and analyze the spread trend of pollutants. They are also unable to quickly predict the spread trend of pollution and plan the inspection route of UAVs based on the prediction results, resulting in low inspection efficiency and failing to give full play to the advantages of UAVs in dynamic monitoring. Based on this, a UAV inspection system for environmental status monitoring is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a drone inspection system for environmental condition monitoring, which solves the technical problems of existing drones being unable to effectively predict and analyze the diffusion trend of pollutants, and having difficulty in quickly predicting the diffusion trend of pollution and planning the inspection route of the drone based on the prediction results.

[0005] A drone inspection system for environmental condition monitoring includes:

[0006] The pollutant monitoring data acquisition module divides the area to be monitored into multiple voxel grids and acquires the pollutant concentration data of each voxel grid at multiple time frames according to the preset time step T. The preset time step T is specifically set to 5 minutes.

[0007] The pollutant monitoring data analysis module obtains the concentration transformation matrix corresponding to each voxel grid based on the pollutant concentration data of each voxel grid in multiple time frames.

[0008] The pollution diffusion point determination module obtains the transformation coefficient of pollutant concentration at each voxel grid with different time frames based on the concentration transformation matrix corresponding to each voxel grid. The pollution diffusion points in the monitoring area are then determined and marked based on the transformation coefficient.

[0009] The pollution diffusion trend line acquisition module analyzes the pollutant concentration data of each voxel grid at multiple time frames, obtains the maximum concentration point corresponding to each time frame, and connects the maximum concentration points of each time frame in sequence according to the order of each time frame to obtain the pollution diffusion trend line of the area to be monitored.

[0010] The inspection point coordinate acquisition module analyzes the pollution diffusion points and pollution diffusion trend lines in the area to be monitored, and obtains the predicted pollution diffusion point coordinates corresponding to the area to be monitored at the next n time steps. The module also analyzes the predicted pollution diffusion point coordinates corresponding to the area to be monitored at the next n time steps to obtain the inspection point coordinates corresponding to the next n time steps, where n is greater than 2.

[0011] The inspection route acquisition module sorts the coordinates of the inspection points corresponding to the next n time steps according to the order of the n time steps, and then obtains the inspection route.

[0012] As a further aspect of the present invention, the specific method for obtaining the concentration transformation matrix corresponding to each voxel grid within the monitoring area is as follows:

[0013] The pollutant concentrations of each voxel grid at different time frames are labeled as Hij. The pollutant concentrations of each voxel grid at different time frames are sorted in the order of the time frames from front to back, thereby obtaining the concentration transformation matrix Ji corresponding to each voxel grid in the area to be monitored. Here, i represents different grids, j represents different time frames, j=1, 2, ..., b, where b represents the total number of time frames, b is a positive integer, and b≥20.

[0014] As a further aspect of the present invention, the specific method for identifying and marking pollution diffusion points in the monitoring area is as follows:

[0015] Through the formula: ; Calculate the transformation coefficient Bi of pollutant concentration at each voxel grid with different time frames, and take the center point of the voxel grid with transformation coefficient Bi greater than the preset value Y1 as the pollution diffusion point Kk in the area to be monitored, where k refers to different pollution diffusion points, k=1, 2, ..., c, where c refers to the total number of pollution diffusions, and c is a positive integer.

[0016] As a further aspect of the present invention, the specific method for obtaining the maximum concentration points corresponding to the monitored area at each time frame is as follows:

[0017] The center point of the voxel grid corresponding to the maximum value of pollutant concentration in each voxel grid at each time frame is taken as the maximum concentration point Zj of the area to be monitored at each time frame. The maximum concentration points Zj of each time frame are connected in sequence according to the corresponding time frame order to obtain the pollution diffusion trend line of the area to be monitored.

[0018] As a further aspect of the present invention, the specific method for obtaining the coordinates of the predicted pollution diffusion points corresponding to the monitored area at n future time steps is as follows:

[0019] Calculate the distance Lr between every two adjacent maximum concentration points. Use the ratio of the sum of all distances Lr to b-1 preset time steps T as the pollution diffusion rate V. Obtain the pollution diffusion direction vector based on the coordinates of the first and last maximum concentration points on the pollution diffusion trend line. The direction vector of pollution diffusion ( x, Convert y) to a unit vector By Skn=(Kxk+V×n×T× ,Kyk+V×n×T× Then, the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at the next n time steps are obtained, where n takes a value greater than 2. Based on the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at the next n time steps, the inspection point coordinates Xn corresponding to the area to be monitored at the next n time steps are obtained, where r represents different distances, r = 1, 2, ..., b-1.

[0020] As a further aspect of the present invention: obtaining the pollution diffusion direction vector. The specific method is as follows:

[0021] Obtain the coordinates Kk(Kxk, Kyk) corresponding to each pollution diffusion point Kk. Mark the coordinates corresponding to the first and last maximum concentration points on the pollution diffusion trend line as F1(F1x, F1y) and F2(F2x, F2y), respectively. Obtain the pollution diffusion direction vector ( x, y).

[0022] As a further aspect of the present invention: the pollution diffusion direction vector Convert to unit vector The specific method is as follows:

[0023] pass, The direction vector of pollution diffusion ( x, Convert y) to a unit vector .

[0024] As a further aspect of the present invention, the specific method for obtaining the coordinates of the inspection points corresponding to the area to be monitored at the next n time steps is as follows:

[0025] The average of the x-coordinate and y-coordinate of the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at each of the next n time steps is used as the x-coordinate and y-coordinate of the corresponding inspection point coordinates Xn at each of the next n time steps, thus obtaining the corresponding inspection point coordinates Xn at each of the next n time steps.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] (1) In this invention, by calculating and comparing the transformation coefficients of all voxel grids with a preset threshold Y1, if the transformation coefficient of a certain grid is greater than the preset value Y1, it is considered that the pollutant concentration in the grid changes abnormally drastically, and there may be a pollution source or the pollution is in the initial stage of diffusion. Therefore, the center point of the voxel grid is marked as the pollution diffusion point. Usually, the accurate location of the active pollution area or the expansion area of ​​the source point is determined by historical data analysis or expert experience, so as to realize the diffusion source determination. By quantifying the dynamic changes of pollutant concentration, it is possible to effectively identify the area where the pollution concentration rises abnormally. These areas are likely to be the source of pollution or the initial point where pollution begins to spread.

[0028] (2) This invention, by constructing a pollution diffusion trend line in the area to be monitored, visually depicts the overall diffusion path and direction of pollutants in space over time, enabling monitoring personnel to quickly grasp the overall movement direction and speed of pollutants, which helps to assess the pollution situation;

[0029] (3) This invention generates an optimal or suboptimal inspection route, enabling the UAV to efficiently pass through these predicted inspection points in sequence, thereby achieving accurate tracking and monitoring of the predicted pollution diffusion route. This ensures that the UAV is always at the forefront or key area of ​​pollution diffusion. By sorting the predicted points and planning the path, an efficient inspection route can be generated, reducing unnecessary detours. This allows the UAV to adapt to changes in the pollution diffusion path in real time, ensuring the effectiveness and timeliness of monitoring. In the event of a sudden pollution incident involving the UAV, the UAV can quickly adjust its inspection route and focus on monitoring the pollution diffusion area. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the system framework structure of the present invention. Detailed Implementation

[0031] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Example 1: Please refer to Figure 1 This application provides an unmanned aerial vehicle (UAV) inspection system for environmental condition monitoring, including:

[0033] The pollutant monitoring data acquisition module divides the area to be monitored into multiple voxel grids. Based on a preset time step T, it acquires the pollutant concentration of different voxel grids within the monitoring area at each time frame. The module then acquires and transmits the pollutant monitoring data of the area to be monitored at each time frame to the pollutant monitoring data analysis module of the UAV, as follows:

[0034] The monitoring area is divided into multiple voxel grids, and the center point of each grid is marked as Gi, where i represents a different grid, i=1,2,...,a, where a represents the total number of grids, a is a positive integer, and a≥1;

[0035] It should be noted that the size of each grid region can be the same or different. Here, the size of each grid region is the same, that is, the monitoring area is uniformly divided into multiple voxel grids of the same size. The determination of the grid size should take into account the monitoring accuracy requirements, the resolution of the UAV sensor, and the limitations of computing resources. For example, for fine pollution source identification, a 1m×1m×1m grid can be used; for large-scale diffusion prediction, the grid size can be appropriately increased.

[0036] According to the preset time step T, the pollutant concentration of different voxel grids in the monitoring area at each time frame is acquired and recorded. The pollutant monitoring data is specifically acquired by setting up pollutant detection devices in different voxel grids in the monitoring area to acquire pollutant concentrations. Pollutant detection devices, including gas sensors, particulate matter sensors, chemical analyzer networks, etc., are strategically deployed inside or near each voxel grid. For grids where fixed sensors cannot be directly deployed, ground robots or tethered drones carrying mobile sensors can be used to take intermittent readings. It should be noted that the preset time step T is specifically set to 5 minutes.

[0037] By dividing the monitoring area into multiple voxel grids and deploying pollutant detection equipment within or near the grids, the area to be monitored can be fully covered, avoiding blind spots and improving the comprehensiveness of monitoring.

[0038] The pollutant monitoring data analysis module is used to acquire and analyze pollutant monitoring data of the monitoring area at various time frames. According to the chronological order of each time frame, the pollutant concentrations of each voxel grid at different time frames are sorted in chronological order from front to back, thereby obtaining the concentration transformation matrix corresponding to each voxel grid in the monitoring area. The specific method is as follows:

[0039] The pollutant concentrations of each voxel grid at different time frames are labeled Hij, where i represents different grids and j represents different time frames, j = 1, 2, ..., b, where b represents the total number of time frames, b is a positive integer, and b ≥ 20. b ≥ 20 is to ensure sufficient time series data for analysis. The pollutant concentrations of each voxel grid at different time frames are sorted in chronological order from front to back, thereby obtaining the concentration transformation matrix Ji (Hi1, Hi2, ..., Hib) corresponding to each voxel grid in the area to be monitored.

[0040] A concentration transformation matrix is ​​constructed by sorting the pollution concentrations in each voxel grid by time. This ensures sufficient time-series data for meaningful analysis, capturing trends and patterns in pollutant concentration changes over time, rather than relying solely on instantaneous values. This matrix is ​​essentially a time-series vector, recording the complete history of pollutant concentration evolution at a specific spatial point—the grid center—over time. This allows the system to move beyond static snapshots and acquire dynamic change analysis capabilities, providing a mathematical foundation for trend identification, abrupt change detection, and anomaly point identification. It also improves the accuracy of pollutant source and diffusion path identification.

[0041] The pollution diffusion point determination module analyzes the concentration of each pollutant in the concentration transformation matrix Ji (Hi1, Hi2, ..., Hib) corresponding to each voxel grid in the monitoring area, thereby obtaining the transformation coefficient of the pollutant concentration at each voxel grid with different time frames. Based on the transformation coefficient at each voxel grid, the pollution diffusion points in the monitoring area are determined and marked. The specific method is as follows:

[0042] Based on the concentration transformation matrix Ji (Hi1, Hi2, ..., Hib) corresponding to each voxel grid in the area to be monitored, the formula is used: ; Calculate the transformation coefficient Bi of pollutant concentration at each voxel grid with different time frames, and take the center point of the voxel grid with transformation coefficient Bi greater than the preset value Y1 as the pollution diffusion point Kk in the area to be monitored, where k refers to different pollution diffusion points, k=1,2,...,c, where c refers to the total number of pollution diffusions, and c is a positive integer.

[0043] It should be noted that the specific value of the preset value Y1 is determined by relevant personnel based on actual needs;

[0044] After calculating the transformation coefficients Bi of all voxel grids, these transformation coefficients Bi are compared with a preset threshold Y1. If the transformation coefficient Bi of a certain grid is greater than Y1, it is considered that the pollutant concentration change within that grid is abnormally drastic, and there may be a pollution source or the beginning stage of pollution diffusion. Therefore, the center point of that voxel grid is marked as the pollution diffusion point Kk. The threshold Y1 can be adjusted according to the actual application scenario, pollutant type, and monitoring accuracy requirements. It is usually determined by historical data analysis or expert experience to accurately locate the active pollution area or the area of ​​source expansion, thus realizing the determination of diffusion source. By quantifying the dynamic changes of pollutant concentration, this module can effectively identify areas with abnormally high pollution concentration. These areas are likely to be the source of pollution or the initial point of pollution diffusion. This is crucial for pollution source tracing and early warning, providing clear key inspection targets for drones, avoiding indiscriminate inspections throughout the area, and saving time and energy.

[0045] The pollution diffusion trend line acquisition module analyzes pollutant monitoring data for the monitored area at various time frames to obtain the maximum concentration points for each time frame. Then, it connects these maximum concentration points sequentially according to the time frame order to obtain the pollution diffusion trend line for the monitored area. The specific method is as follows:

[0046] The center point of the voxel grid corresponding to the maximum value of pollutant concentration in each voxel grid at each time frame is taken as the maximum concentration point Zj of the area to be monitored at each time frame. The maximum concentration points Zj of each time frame are connected in the order of the corresponding time frames to obtain the pollution diffusion trend line Q of the area to be monitored.

[0047] After obtaining the coordinates of all maximum concentration points from the first time frame to the b-th time frame, the system connects these maximum concentration points sequentially according to time. These connecting lines together form the pollution diffusion trend line of the monitored area. This trend line intuitively depicts the overall diffusion path and direction of pollutants in space over time. The pollution diffusion trend line provides a highly visualized and easy-to-understand pollution diffusion trajectory, enabling monitoring personnel to quickly grasp the overall movement direction and speed of pollutants, which helps to assess the pollution situation.

[0048] Abstracting a large amount of grid concentration data into a core trend line reduces the complexity of data interpretation and makes the pollution diffusion pattern clearer. By observing the trend line, managers can make preliminary estimates of the scope and impact of pollution diffusion, providing important decision support for emergency response, resource allocation and pollution control measures.

[0049] The inspection point coordinate acquisition module analyzes the pollution diffusion points and pollution diffusion trend lines in the area to be monitored, and obtains the predicted pollution diffusion point coordinates for the area to be monitored at each of the next n time steps. Based on the predicted pollution diffusion point coordinates for the area to be monitored at each of the next n time steps, the corresponding inspection point coordinates for each of the next n time steps are obtained, as follows:

[0050] The coordinates of the maximum concentration point Zj at each time frame are labeled as Zj(Zxj, Zyj). The distance Lr between any two adjacent maximum concentration points is calculated using the distance formula, where r represents different distances, r = 1, 2, ..., b-1. The ratio of the sum of all distances Lr to b-1 preset time steps T is taken as the pollution diffusion rate V, i.e., using the formula: The rate of pollution diffusion, V, is obtained. This rate V reflects the average diffusion speed of the pollutant during the monitoring period and is a key parameter for predicting future diffusion.

[0051] Obtain the coordinates Kk(Kxk, Kyk) corresponding to each pollution diffusion point Kk, and obtain the coordinates F1(F1x, F1y) and F2(F2x, F2y) corresponding to the first and last maximum concentration points on the pollution diffusion trend line, respectively. Obtain the pollution diffusion direction vector ( x, y), this vector represents the overall direction of pollution diffusion;

[0052] pass, The direction vector of pollution diffusion ( x, Convert y) to a unit vector The length of the vector is normalized to 1, and only the direction information is retained for the unit vector, which facilitates subsequent position calculation based on velocity and time step.

[0053] By Skn=(Kxk+V×n×T× ,Kyk+V×n×T× ), thereby obtaining the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at n future time steps, where the specific value of n is greater than 2;

[0054] Based on the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at the next n time steps, the coordinates Xn of the inspection points corresponding to the area to be monitored at the next n time steps are obtained. The specific method is as follows:

[0055] The average of the x-coordinates and y-coordinates of the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at the next n time steps is used as the x-coordinates and y-coordinates of the corresponding inspection point coordinates Xn at the next n time steps. Thus, the corresponding inspection point coordinates Xn at the next n time steps are obtained. The Xn obtained in this way represents the center of the area that all known pollution diffusion points may spread to at a certain time point in the future. It is the best position for UAV to carry out concentrated inspection. By averaging multiple diffusion points, it can cope with multi-source diffusion or complex terrain to a certain extent and make the inspection points more representative.

[0056] The analysis of historical data is transformed into predictions of future pollution trends, enabling drones to be deployed in advance to areas where pollutants are likely to reach, rather than passively chasing pollution. Through predicted inspection points, drones can focus their efforts on monitoring pollution hotspots, significantly improving inspection efficiency and coverage of key areas, reducing the ineffective flight time of drones in non-polluted areas, extending the drones' endurance, and lowering operating costs.

[0057] The inspection route acquisition module sorts the coordinates Xn of the inspection points at n future time steps according to the order of the n time steps to obtain the inspection route, and outputs it to the UAV control unit to control the UAV's inspection route.

[0058] By generating an optimal or suboptimal inspection route, drones can efficiently pass through these predicted inspection points sequentially, achieving accurate tracking and monitoring of predicted pollution spread paths. This dynamic adjustment mechanism ensures that drones remain at the forefront or key areas of pollution spread, directly translating data analysis and prediction results into specific drone flight missions. This significantly enhances the autonomy and intelligence of drones. By sorting and planning the predicted points, efficient inspection routes can be generated, reducing unnecessary detours, thereby saving energy and extending the drone's operating time. This allows drones to adapt to changes in pollution spread paths in real time, ensuring the effectiveness and timeliness of monitoring. In the event of a sudden pollution incident, drones can quickly adjust their inspection routes to focus on monitoring the pollution spread areas.

[0059] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0060] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A drone inspection system for environmental condition monitoring, characterized in that, include: The pollutant monitoring data acquisition module divides the area to be monitored into multiple voxel grids and acquires the pollutant concentration data of each voxel grid at multiple time frames according to the preset time step T. The preset time step T is specifically set to 5 minutes. The pollutant monitoring data analysis module obtains the concentration transformation matrix corresponding to each voxel grid based on the pollutant concentration data of each voxel grid in multiple time frames. The pollution diffusion point determination module obtains the transformation coefficient of pollutant concentration at each voxel grid with different time frames based on the concentration transformation matrix corresponding to each voxel grid. The pollution diffusion points in the monitoring area are then determined and marked based on the transformation coefficient. The pollution diffusion trend line acquisition module analyzes the pollutant concentration data of each voxel grid at multiple time frames, obtains the maximum concentration point corresponding to each time frame, and connects the maximum concentration points of each time frame in sequence according to the order of each time frame to obtain the pollution diffusion trend line of the area to be monitored. The inspection point coordinate acquisition module analyzes the pollution diffusion points and pollution diffusion trend lines in the area to be monitored, obtaining the predicted pollution diffusion point coordinates for each of the next n time steps. This process is repeated to obtain the corresponding inspection point coordinates for each of the next n time steps, where n is greater than 2. The inspection route acquisition module sorts the coordinates of the inspection points corresponding to the next n time steps according to the order of the n time steps, and then obtains the inspection route. The specific method for obtaining the concentration transformation matrix corresponding to each voxel grid within the monitored area is as follows: The pollutant concentrations of each voxel grid at different time frames are labeled as Hij. The pollutant concentrations of each voxel grid at different time frames are sorted in the order of time frames from front to back, thereby obtaining the concentration transformation matrix Ji corresponding to each voxel grid in the area to be monitored. Here, i represents different grids, j represents different time frames, j=1, 2, ..., b, where b represents the total number of time frames, b is a positive integer, and b≥20. The specific method for identifying and marking pollution diffusion points in the monitoring area is as follows: Through the formula: ; Calculate the transformation coefficient Bi of pollutant concentration at each voxel grid with different time frames, and take the center point of the voxel grid with transformation coefficient Bi greater than the preset value Y1 as the pollution diffusion point Kk in the area to be monitored, where k refers to different pollution diffusion points, k=1, 2, ..., c, where c refers to the total number of pollution diffusions, and c is a positive integer.

2. The UAV inspection system for environmental condition monitoring according to claim 1, characterized in that, The specific method for obtaining the maximum concentration points of the monitored area at each time frame is as follows: The center point of the voxel grid corresponding to the maximum value of pollutant concentration in each voxel grid at each time frame is taken as the maximum concentration point Zj of the area to be monitored at each time frame. The maximum concentration points Zj of each time frame are connected in sequence according to the corresponding time frame order to obtain the pollution diffusion trend line of the area to be monitored.

3. The UAV inspection system for environmental condition monitoring according to claim 2, characterized in that, The specific method for obtaining the coordinates of the predicted pollution diffusion points for the monitored area at n future time steps is as follows: Calculate the distance Lr between every two adjacent maximum concentration points. Use the ratio of the sum of all distances Lr to b-1 preset time steps T as the pollution diffusion rate V. Obtain the pollution diffusion direction vector based on the coordinates of the first and last maximum concentration points on the pollution diffusion trend line. The direction vector of pollution diffusion ( x, Convert y) to a unit vector By Skn=(Kxk+V×n×T× ,Kyk+V×n×T× Then, the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at the next n time steps are obtained, where n is greater than 2. Based on the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at the next n time steps, the inspection point coordinates Xn corresponding to the area to be monitored at the next n time steps are obtained, where r represents different distances, r = 1, 2, ..., b-1.

4. The UAV inspection system for environmental condition monitoring according to claim 3, characterized in that, Obtain the direction vector of pollution diffusion The specific method is as follows: Obtain the coordinates Kk(Kxk, Kyk) corresponding to each pollution diffusion point Kk. Mark the coordinates corresponding to the first and last maximum concentration points on the pollution diffusion trend line as F1(F1x, F1y) and F2(F2x, F2y), respectively. Obtain the pollution diffusion direction vector ( x, y).

5. The UAV inspection system for environmental condition monitoring according to claim 4, characterized in that, The direction vector of pollution diffusion Convert to unit vector The specific method is as follows: pass, The direction vector of pollution diffusion ( x, Convert y) to a unit vector .

6. The UAV inspection system for environmental condition monitoring according to claim 3, characterized in that, The specific method for obtaining the coordinates of the inspection points corresponding to the monitored area at the next n time steps is as follows: The average of the x-coordinate and y-coordinate of the predicted pollution diffusion point coordinates Skn corresponding to each pollution diffusion point Kk at each of the next n time steps is used as the x-coordinate and y-coordinate of the corresponding inspection point coordinates Xn at each of the next n time steps, thus obtaining the corresponding inspection point coordinates Xn at each of the next n time steps.

Images