Atmospheric environment monitoring data space modeling method based on unmanned aerial vehicle platform

Abstract

The invention provides an atmospheric environment monitoring data space modeling method based on an unmanned aerial vehicle platform. S1, collecting a certain amount of scattered atmosphere data through the unmanned aerial vehicle, recording the scattered atmosphere data and corresponding coordinates, and transmitting the data and the coordinates to the cloud; s2, estimating atmosphere data of unmeasured coordinate points based on the atmosphere data of the measured coordinate points by using a Kriging interpolation technology; s3, calculating an estimated deviation value of the atmosphere data of the unmeasured coordinate points; s4, driving the unmanned aerial vehicle to acquire data at the coordinate point with the maximum estimated deviation value and transmitting the data to the cloud; s5, recalculating the estimated deviation value of each unmeasured coordinate point based on the newly collected data; and S6, repeating the steps S4 to S5 until the maximum value of the estimated deviation value of each unmeasured coordinate point in the space is smaller than a set allowable value. The method not only constructs a high-precision atmospheric data space model, but also provides a scientific basis for adjusting and optimizing the data acquisition trajectory of the unmanned aerial vehicle in real time by calculating the estimation deviation value of the estimation point.

Description

Technical Field

[0001] The present invention relates to the field of environmental monitoring, and in particular to a method for spatial modeling of atmospheric environmental monitoring data based on an unmanned aerial vehicle platform. Background Art

[0002] Traditional atmospheric environmental monitoring mainly relies on fixed monitoring stations set up on the ground for data collection. Fixed monitoring stations monitor pollutants in the air such as sulfur dioxide, nitrogen oxides, carbon monoxide, particulate matter (PM2.5, PM10, etc.) and volatile organic compounds at specific locations.

[0003] Although the above-mentioned fixed monitoring stations set up on the ground can provide relatively accurate data, they are often difficult to fully cover a wide geographical area due to the number and layout of the stations. In particular, in areas with complex terrain and changing environments, the spatial coverage of monitoring data is obviously insufficient. Although spatial data modeling technology can construct a spatial distribution model of the atmospheric environment based on data from fixed monitoring stations, since fixed stations can only collect data at fixed coordinates, it is impossible to dynamically adjust the most effective data collection points according to changes in the atmospheric environment, thus affecting the accuracy and real-time performance of spatial modeling. In addition, the high cost of building environmental monitoring stations at a high density leads to sparse distribution of monitoring stations, which further limits the accuracy and detail of spatial modeling of environmental data, and also brings challenges to spatial modeling of environmental data in local areas.

[0004] Based on this, the present application proposes a method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform. In atmospheric environment monitoring, the method overcomes the problem that traditional fixed monitoring sites are limited in number and layout and are difficult to fully cover geographical areas; at the same time, the method overcomes the problem that when data spatial modeling is performed based on monitoring station data, the collection points cannot be dynamically adjusted, resulting in insufficient spatial modeling accuracy and real-time performance; and the method overcomes the problem that the sparse distribution of monitoring stations due to high construction costs limits the accuracy and detail of spatial modeling of environmental data. Summary of the invention

[0005] The purpose of the present invention is to propose a spatial modeling method for atmospheric environment monitoring data based on an unmanned aerial vehicle platform, which utilizes the unmanned aerial vehicle platform equipped with monitoring sensor equipment and a satellite navigation system, combined with a cloud computing platform, to achieve efficient, flexible and accurate spatial modeling of atmospheric environment monitoring data, and improve the modeling accuracy by dynamically optimizing the data acquisition trajectory, thereby effectively solving the above three technical problems.

[0006] To achieve the above-mentioned purpose, the technical solution of the present invention is: a method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform, characterized in that the equipment relied on by the method includes a monitoring sensor device carried by the unmanned aerial vehicle, a satellite navigation signal receiving device and a cloud computing platform, and includes the following steps;

[0007] S1: Collect a certain amount of scattered atmospheric data through drones, record the scattered atmospheric data and their corresponding coordinates, and transmit them to the cloud;

[0008] S2: Use Kriging interpolation technology to estimate the atmospheric data of unmeasured coordinate points based on the atmospheric data of measured coordinate points;

[0009] S3: Calculate the estimated deviation value of the atmospheric data of the unmeasured coordinate point;

[0010] S4: Use satellite navigation to drive the drone to the coordinate point with the largest estimated deviation value to collect data and transmit it to the cloud;

[0011] S5: Based on the newly collected data, the atmospheric data space model is calculated and updated according to steps S3 and S4, and the estimated deviation value of each unmeasured coordinate point is recalculated;

[0012] S6: Repeat steps S4 to S5 until the maximum value of the estimated deviation value of each unmeasured coordinate point in the space is less than the set allowable value.

[0013] Preferably, the monitoring objects of the atmospheric data cover a variety of pollutants in the air, including sulfur dioxide, nitrogen oxides, carbon monoxide, particulate matter and volatile organic compounds.

[0014] Preferably, step S1 specifically comprises: collecting a certain amount of scattered atmospheric data by carrying monitoring sensor equipment and satellite navigation signal receiving equipment on a drone, and storing the data in the form of a vector z; and recording the coordinates corresponding to the data, and storing the data in the form of a matrix X; and transmitting the vector z and the corresponding coordinate matrix X to the cloud through the Internet of Things communication technology.

[0015] Preferably, the IoT communication technology includes any one of 4G technology, 5G technology, and NB-IoT technology.

[0016] Preferably, the Kriging interpolation technique used includes any one of ordinary Kriging interpolation, regression Kriging interpolation and co-Kriging interpolation.

[0017] Preferably, the kriging interpolation technique used is ordinary kriging interpolation, and the method of estimating the atmospheric data of the unmeasured coordinate points based on the atmospheric data of the measured coordinate points is specifically as follows:

[0018] S2.1: Construct the spatial autocorrelation function between spatial data;

[0019] S2.2: Calculate the covariance matrix V0 between the measured coordinate points and the covariance matrix V between the unmeasured coordinate points and the measured coordinate points c ;

[0020]

[0021] In the formula, ρ represents the autocorrelation coefficient, which is calculated by the spatial autocorrelation function; d represents the number of measured data points; n represents the number of unmeasured data points; σ represents the standard deviation, which is obtained from the measured atmospheric data;

[0022] S2.3: Calculate the interpolation weights and perform interpolation calculations on the unmeasured coordinates:

[0023]

[0024] z kr,k =z T β (5)

[0025] Where l is a matrix whose elements are all 1; β is a vector containing interpolation weight coefficients; λ is the Lagrange multiplier; k is the kth unmeasured point; V c (k) is the kth column of the covariance matrix Vc, describing the correlation between all measured data points and the kth unmeasured point, z kr,k Represents the estimated value of atmospheric data for the kth unmeasured point.

[0026] Preferably, the spatial autocorrelation function adopts a theoretical spatial autocorrelation function, including any one of an exponential function, a Gaussian function, and a Matérn equation.

[0027] Preferably, the spatial autocorrelation function adopts the form of a Gaussian function, specifically:

[0028]

[0029] In the formula, τ x , τ y and τ z Respectively represent the distance between any two points in the space in the x, y and z directions, calculated using the coordinate information stored in the coordinate matrix X; θ x ,θ y and θ z Represent the autocorrelation distance parameters in the x, y and z directions respectively.

[0030] Preferably, the estimated deviation value of the atmospheric data of the unmeasured coordinate point is calculated as follows:

[0031] σ e =(V0 -1 V c ) T V0 (6)

[0032] In the formula, σ e is a vector containing the estimated deviations for each unmeasured point.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention relies on the monitoring sensor equipment and satellite navigation system carried by the drone to provide a high-efficiency and high-flexibility dynamic data collection method in three-dimensional space for the spatial modeling of atmospheric data; at the same time, through the linkage mechanism between the drone equipment and the cloud computing platform, real-time calculation and dynamic optimization of data collection trajectory are realized. More importantly, by applying the Kriging interpolation technology, this method not only constructs a high-precision atmospheric data spatial model, but also provides a scientific basis for real-time adjustment and optimization of the drone data collection trajectory by calculating the estimated deviation value of the estimated point. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0036] The following is combined with Figure 1 , the technical solution of the present invention is specifically described.

[0037] The present invention proposes a method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform, and the equipment and devices relied on by the method include monitoring sensor equipment carried by the unmanned aerial vehicle, satellite navigation signal receiving equipment and a cloud computing platform. The present invention does not specifically limit the monitoring sensors and satellite navigation signal receiving equipment carried by the unmanned aerial vehicle. This method also does not limit the monitoring objects of atmospheric data, which generally cover a variety of pollutants in the air, such as sulfur dioxide, nitrogen oxides, carbon monoxide, particulate matter (PM2.5, PM10, etc.) and volatile organic compounds. The data analysis and calculation process in the method of the present invention is carried out in a cloud computing platform (i.e., the cloud), and the type of cloud computing platform is not limited. The specific steps of the present invention include:

[0038] S1: Collect a certain amount of scattered atmospheric data through drones, record the scattered atmospheric data and their corresponding coordinates, and transmit them to the cloud;

[0039] S2: Using Kriging interpolation technology, based on the atmospheric data of the measured coordinate points, the atmospheric data of the unmeasured coordinate points are estimated. The present invention does not limit the type of Kriging interpolation technology, and common ones include ordinary Kriging interpolation, regression Kriging interpolation, and co-Kriging interpolation;

[0040] S3: Calculate the estimated deviation value of the atmospheric data of the unmeasured coordinate point;

[0041] S4: Use satellite navigation to drive the drone to the coordinate point with the largest estimated deviation value to collect data and transmit it to the cloud;

[0042] S5 calculates and updates the atmospheric data spatial model based on the newly collected data, and recalculates the estimated deviation value of each unmeasured coordinate point;

[0043] S6: Repeat steps S4 to S5 until the maximum value of the estimated deviation value of each unmeasured coordinate point in the space is less than the set allowable value, which can be flexibly set according to the specific project conditions.

[0044] The above steps are described in detail below.

[0045] S1: A certain amount of scattered atmospheric data is collected by using drones equipped with monitoring sensors and satellite navigation signal receiving equipment, stored in the form of vector z, and the coordinates corresponding to the data are recorded and stored in the form of matrix X. The collected data is transmitted to the cloud through IoT communication technologies such as 4G, 5G, NB-IoT, etc.

[0046] S2: In the cloud computing platform, the Kriging interpolation technology is used to estimate the atmospheric data of the unmeasured coordinate points based on the atmospheric data of the measured coordinate points. The present invention does not limit the type of Kriging interpolation technology. Common ones include ordinary Kriging interpolation, regression Kriging interpolation, and collaborative Kriging interpolation. The following takes the calculation method of ordinary Kriging interpolation as an example to explain the method of calculating Kriging interpolation. The main principles of other Kriging interpolation calculations are similar to ordinary Kriging interpolation:

[0047] 1) Calculate the autocorrelation function between spatial data. The spatial autocorrelation function is a function used to describe the autocorrelation relationship of spatial data. The main parameter of the spatial autocorrelation function is: spatial autocorrelation distance. Generally speaking, the larger the spatial autocorrelation distance, the stronger the autocorrelation of spatial data. The form of the autocorrelation function can be selected from commonly used theoretical autocorrelation functions, such as exponential function and Gaussian function. You can also choose the Matérn equation with a flexible function form.

[0048] The following is an explanation of the calculation steps using the Gaussian function as an example:

[0049]

[0050] In formula (1), τ x , τ y and τ z Respectively represent the distance between any two points in the space in the x, y and z directions, and are calculated using the coordinate information stored in the coordinate matrix X. The coordinates of the unmeasured points are determined by the user dividing the space into grids and defining a three-dimensional Cartesian coordinate system. This space area is determined by the user according to his or her own needs. The coordinates determined by the drone through the satellite can be converted to the divided coordinate grid. x ,θ y and θ zThey represent the autocorrelation distance parameters in the x, y and z directions respectively, and can be solved by a parameter optimization algorithm according to the collected spatial coordinate data and the atmospheric data corresponding to each coordinate, such as the maximum likelihood estimation method.

[0051] 2) Calculate the covariance matrix V0 between the measured coordinate points and the covariance matrix V between the unmeasured coordinate points and the measured coordinate points c .

[0052]

[0053] In the above formula, ρ represents the autocorrelation coefficient obtained by formula (1); d represents the number of measured data points; n represents the number of unmeasured data points; σ represents the standard deviation, which is obtained from the measured atmospheric data.

[0054] 3) Calculate the interpolation weight and perform interpolation calculation on the unmeasured coordinates:

[0055]

[0056] z kr,k =z T β (5)

[0057] In the above formula, l represents a matrix whose elements are all 1; β represents a vector containing interpolation weight coefficients; λ is the Lagrange multiplier; k represents the kth unmeasured point; V c (k) is the kth column of the covariance matrix Vc, describing the correlation between all measured data points and the kth unmeasured point, z kr,k Represents the estimated value of atmospheric data for the kth unmeasured point.

[0058] S3: Calculate the estimated deviation value of the atmospheric data of the unmeasured coordinate point:

[0059] σ e =(V0 -1 V c ) T V0 (6)

[0060] In formula 6, σ e is a vector containing the estimated deviations for each unmeasured point.

[0061] S4: Based on the calculated data in step (3), the drone is driven to the coordinate point with the largest estimated deviation value to collect data and transmit it to the cloud. This process can be achieved through satellite navigation equipment and the automatic cruise function of the drone.

[0062] S5: Based on the newly collected data, the atmospheric data space model is calculated and updated, and the estimated deviation value of each unmeasured coordinate point is recalculated. The calculation method is as shown in steps (3) and (4).

[0063] S6: Repeat steps (4) to (5) until the maximum value of the estimated deviation value of each unmeasured coordinate point in the space is less than the set allowable value, which can be flexibly set according to the specific project conditions.

[0064] The above are preferred embodiments of the present invention. Any changes made according to the technical solution of the present invention, as long as the resulting functions do not exceed the scope of the technical solution of the present invention, belong to the protection scope of the present invention.

Claims

1. A spatial modeling method for atmospheric environment monitoring data based on an unmanned aerial vehicle platform, characterized in that: The method relies on equipment including monitoring sensor equipment carried by the UAV, satellite navigation signal receiving equipment and a cloud computing platform, and includes the following steps: S1: Collect a certain amount of scattered atmospheric data through drones, record the scattered atmospheric data and their corresponding coordinates, and transmit them to the cloud; S2: Use Kriging interpolation technology to estimate the atmospheric data of unmeasured coordinate points based on the atmospheric data of measured coordinate points; S3: Calculate the estimated deviation value of the atmospheric data of the unmeasured coordinate point; S4: Use satellite navigation to drive the drone to the coordinate point with the largest estimated deviation value to collect data and transmit it to the cloud; S5: Based on the newly collected data, the atmospheric data space model is calculated and updated according to steps S3 and S4, and the estimated deviation value of each unmeasured coordinate point is recalculated; S6: Repeat steps S4 to S5 until the maximum value of the estimated deviation value of each unmeasured coordinate point in the space is less than the set allowable value.

2. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 1 is characterized in that: The monitoring objects of the atmospheric data cover a variety of pollutants in the air, including sulfur dioxide, nitrogen oxides, carbon monoxide, particulate matter and volatile organic compounds.

3. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 1 is characterized in that: The step S1 is specifically as follows: using a drone equipped with monitoring sensor equipment and a satellite navigation signal receiving device to collect a certain amount of scattered atmospheric data and store it in the form of a vector z; and recording the coordinates corresponding to the data and storing them in the form of a matrix X; and transmitting the vector z and the corresponding coordinate matrix X to the cloud through the Internet of Things communication technology.

4. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 3 is characterized in that: The Internet of Things communication technology includes any one of 4G technology, 5G technology, and NB-IoT technology.

5. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 1, characterized in that: The kriging interpolation techniques used include ordinary kriging interpolation, regression kriging interpolation, and cokriging interpolation.

6. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 3 is characterized in that: The Kriging interpolation technique used is ordinary Kriging interpolation. The atmospheric data of the unmeasured coordinate points are estimated based on the atmospheric data of the measured coordinate points as follows: S2.1: Construct the spatial autocorrelation function between spatial data; S2.2: Calculate the covariance matrix V0 between the measured coordinate points and the covariance matrix V between the unmeasured coordinate points and the measured coordinate points c ; In the formula, ρ represents the autocorrelation coefficient, which is calculated by the spatial autocorrelation function; d represents the number of measured data points; n represents the number of unmeasured data points; σ represents the standard deviation, which is obtained from the measured atmospheric data; S2.3: Calculate the interpolation weights and perform interpolation calculations on the unmeasured coordinates: With kr,k =from T β (5) Where l is a matrix whose elements are all 1; β is a vector containing interpolation weight coefficients; λ is the Lagrange multiplier; k is the kth unmeasured point; V c (k) is the kth column of the covariance matrix Vc, describing the correlation between all measured data points and the kth unmeasured point, z kr,k Represents the estimated value of atmospheric data for the kth unmeasured point.

7. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 6 is characterized in that: The spatial autocorrelation function adopts a theoretical spatial autocorrelation function, including any one of an exponential function, a Gaussian function, and a Matérn equation.

8. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 6 is characterized in that: The spatial autocorrelation function adopts the form of a Gaussian function, specifically: In the formula, τ x , τ y and τ z Respectively represent the distance between any two points in the space in the x, y and z directions, calculated using the coordinate information stored in the coordinate matrix X; θ x ,θ y and θ z Represent the autocorrelation distance parameters in the x, y and z directions respectively.

9. The method for spatial modeling of atmospheric environment monitoring data based on an unmanned aerial vehicle platform according to claim 6, characterized in that: The estimated deviation value of the atmospheric data of the unmeasured coordinate point is calculated as follows: σ e =(V0 -1 V c ) T V0 (6) In the formula, σ e is a vector containing the estimated deviations for each unmeasured point.

Images