Rapid source tracing method and system for pollution source based on dual-unmanned aerial vehicle cooperative gradient detection

Abstract

The invention provides a rapid source tracing method and system for a pollution source based on dual unmanned aerial vehicle cooperative gradient detection, and belongs to the technical field of environment monitoring and unmanned aerial vehicle application, and the method comprises the steps: initializing system parameters; deploying the first unmanned aerial vehicle to hover at an initial position, and measuring an atmospheric pollutant concentration value at the initial position and recording the atmospheric pollutant concentration value as a first concentration value; the second unmanned aerial vehicle is controlled to fly around, and the pollutant concentration value is measured and recorded as a second concentration value; calculating a pollutant concentration gradient value; based on the concentration gradient value, the direction with the maximum atmospheric pollutant concentration gradient in the whole-cycle range of 0-360 degrees is determined and recorded as the maximum gradient direction; the second unmanned aerial vehicle is controlled to move to the intersection point position of the maximum gradient direction and the fly-around circumference for hovering; the first unmanned aerial vehicle is controlled to fly around, and the concentration gradient measurement and maximum gradient direction determination process is repeated; and alternately executing the operation. Through the working mode that the two unmanned aerial vehicles alternately fly around and hover, rapid and efficient pollution source positioning is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of environmental monitoring and unmanned aerial vehicle application technology, in particular to a pollution source rapid tracing method and system based on double unmanned aerial vehicle cooperative gradient detection. BACKGROUND

[0002] Traditional pollution source tracing methods mainly rely on fixed monitoring station networks or single-person carrying equipment to conduct regional investigation, which has problems such as low efficiency, high cost, and limited coverage.

[0003] In recent years, unmanned aerial vehicle technology has been gradually applied to the field of environmental monitoring due to its high flexibility and relatively low cost. In the prior art, there are schemes for using a single unmanned aerial vehicle to trace pollution, but the single unmanned aerial vehicle scheme usually adopts a zigzag or grid path search, which is low in efficiency and easy to fall into a local optimal solution.

[0004] In addition, existing multi-unmanned aerial vehicle cooperative schemes mostly use swarm intelligence algorithms, which require a large number of unmanned aerial vehicles to work simultaneously, resulting in high cost and complex coordination. Therefore, there is a need for a pollution source tracing solution that can improve tracing efficiency without significantly increasing cost. SUMMARY

[0005] The present application provides a pollution source rapid tracing method and system based on double unmanned aerial vehicle cooperative gradient detection, which realizes rapid and efficient pollution source positioning through the alternating flight and hovering modes of two unmanned aerial vehicles.

[0006] One aspect of the present specification discloses a pollution source rapid tracing method based on double unmanned aerial vehicle cooperative gradient detection, comprising:

[0007] S101: initializing system parameters, the system parameters including a preset radius of unmanned aerial vehicle flight, a number of equiangular interval points during unmanned aerial vehicle flight, a concentration gradient threshold for determining the termination of tracing, and a unmanned aerial vehicle position change setting value for determining the convergence of tracing;

[0008] S102: deploying a first unmanned aerial vehicle to hover at an initial position, measuring the atmospheric pollutant concentration value of the initial position by an atmospheric pollution sensor carried by the first unmanned aerial vehicle, and recording it as a first concentration value;

[0009] S103: controlling a second unmanned aerial vehicle to fly around the first unmanned aerial vehicle with a preset radius as the circumference, and stopping at multiple equiangular interval points, measuring the atmospheric pollutant concentration value of each stopping point by an atmospheric pollution sensor carried by the second unmanned aerial vehicle, and recording it as a second concentration value;

[0010] S104: calculating the atmospheric pollutant concentration gradient value of the second unmanned aerial vehicle at each equiangular interval point from the first unmanned aerial vehicle according to the first concentration value, the second concentration value, and the preset radius;

[0011] S105: determining a direction with the largest concentration gradient of the atmospheric pollutants in the full range of 0°-360° based on the concentration gradient values calculated in S104, denoted as the maximum gradient direction;

[0012] S106: controlling the second unmanned aerial vehicle to move to the intersection position of the maximum gradient direction and the circumferential flight in S103, and hovering at the intersection position;

[0013] S107: controlling the first unmanned aerial vehicle to circumferentially fly around the second unmanned aerial vehicle hovering in S106 with a preset radius as the circumferential flight, and stopping at a plurality of equiangular interval points to measure the atmospheric pollutant concentration values at each stop point, repeating the concentration gradient measurement and maximum gradient direction determination processes of S104-S105;

[0014] S108: alternately performing the operations of S103-S107, i.e. sequentially performing concentration measurement, concentration gradient value calculation and maximum gradient direction determination with the currently hovering unmanned aerial vehicle as the center and the other unmanned aerial vehicle as the circumferential flight unmanned aerial vehicle, and controlling the circumferential flight unmanned aerial vehicle to move to the intersection of the maximum gradient direction and the circumferential flight;

[0015] S109: after each alternately performing S103-S108, determining whether the tracing termination condition is met according to the concentration gradient threshold value and the unmanned aerial vehicle position change setting value, if met, terminating the tracing process, if not met, continuing to alternate.

[0016] In the present specification, the preset radius of the circumferential flight of the unmanned aerial vehicle is in the range of 30 meters to 100 meters, which is determined according to the terrain complexity of the monitoring area and the atmospheric pollutant diffusion speed. The flatter the terrain and the faster the diffusion speed, the larger the preset radius value.

[0017] In the present specification, the number of equiangular interval points when the second unmanned aerial vehicle circumferentially flies is at least 8, and the angle difference between each equiangular interval point is equal, and the accuracy of the concentration gradient detection in the full range of 0°-360° is improved by increasing the number of equiangular interval points.

[0018] In the present specification, when calculating the concentration gradient value, the absolute value difference between the second concentration value and the first concentration value at each equiangular interval point is calculated based on the first concentration value of the first unmanned aerial vehicle, and then the absolute value difference is divided by the preset radius to obtain the concentration gradient value corresponding to the equiangular interval point.

[0019] In the present specification, the specific process of determining the maximum gradient direction is as follows: first, the concentration gradient values of each equiangular interval point are processed by using an interpolation algorithm to construct a complete atmospheric pollutant concentration gradient distribution curve in the range of 0°-360°, and then the gradient distribution curve is analyzed to find the extreme point in the curve, and the direction corresponding to the extreme point is the maximum gradient direction in the full range of 0°-360°.

[0020] In the specification, in the flying-around of the first unmanned aerial vehicle with the second unmanned aerial vehicle as the center in S107, the flying-around operation of the second unmanned aerial vehicle in S103 needs to be consistent, that is, the same preset radius and equal-angle interval point quantity are adopted, and the atmospheric pollutant concentration value is measured at each equal-angle interval point to ensure the uniformity of the parameters of the two flying-around measurements to improve the gradient calculation accuracy.

[0021] In the specification, in the alternately performing operation in S108, after the unmanned aerial vehicle is moved to the intersection position and hovers each time, the accuracy of the hovering position needs to be confirmed by the positioning device carried by the unmanned aerial vehicle. If the deviation of the hovering position from the intersection position exceeds the preset accuracy value, the position of the unmanned aerial vehicle is adjusted to meet the accuracy requirement before the next flying-around operation is performed.

[0022] In the specification, the specific way of judging the termination condition of the tracing source in S109 is: if the currently calculated concentration gradient value is lower than the concentration gradient threshold value, it is determined that the vicinity of the pollution source has been reached; if the position change of the unmanned aerial vehicle in the last 3 iterations is less than the set value of the position change of the unmanned aerial vehicle, it is determined that the tracing process converges; any one of the above conditions is met, and the tracing is terminated.

[0023] In the specification, in the flying-around process of the unmanned aerial vehicle in S103 and S107, the flight speed needs to be kept stable, and the deviation of the flying-around track from the circle with the hovering unmanned aerial vehicle as the center and the preset radius as the radius is not more than 5%, so as to avoid the change of the distance between the two unmanned aerial vehicles due to the deviation of the track, and affect the accuracy of the concentration gradient calculation.

[0024] Another aspect of the specification discloses a pollution source rapid tracing system based on double unmanned aerial vehicle cooperative gradient detection, which applies the pollution source rapid tracing method based on double unmanned aerial vehicle cooperative gradient detection in any one of the above aspects. The pollution source rapid tracing system based on double unmanned aerial vehicle cooperative gradient detection comprises:

[0025] An initialization module is configured to initialize system parameters, including a preset radius of the flying-around of the unmanned aerial vehicle, an equal-angle interval point quantity during the flying-around of the unmanned aerial vehicle, a concentration gradient threshold value for judging the termination of the tracing, and a set value of the position change of the unmanned aerial vehicle for judging the convergence of the tracing;

[0026] A first measuring module is configured to arrange the first unmanned aerial vehicle to hover at an initial position, and measure the atmospheric pollutant concentration value of the initial position by the atmospheric pollution sensor carried by the first unmanned aerial vehicle, which is recorded as a first concentration value;

[0027] A second measuring module is configured to control the second unmanned aerial vehicle to fly around with the first unmanned aerial vehicle as the center and a preset radius as the circle, and stop at a plurality of equal-angle interval points, and measure the atmospheric pollutant concentration value of each stopping point by the atmospheric pollution sensor carried by the second unmanned aerial vehicle, which is recorded as a second concentration value;

[0028] The first calculation module is configured to calculate, according to the first concentration value, the second concentration value, and the preset radius, the atmospheric pollutant concentration gradient value of the second unmanned aerial vehicle at each equiangular interval point relative to the first unmanned aerial vehicle;

[0029] The second calculation module is configured to determine, based on the concentration gradient values calculated by the first calculation module, a direction with the maximum atmospheric pollutant concentration gradient within a full range of 0°-360°, denoted as a maximum gradient direction.

[0030] The moving module is configured to control the second unmanned aerial vehicle to move to an intersection position of the maximum gradient direction and a flight circle of the second measurement module, and to hover at the intersection position.

[0031] The direction determination module is configured to control the first unmanned aerial vehicle to fly around the second unmanned aerial vehicle hovering in the moving module as a center and a preset radius as a circle, to stop at each equiangular interval point and measure the atmospheric pollutant concentration value at each stop point, and to repeat the concentration gradient measurement and the maximum gradient direction determination process.

[0032] The alternating execution module is configured to alternately execute the operations of: sequentially performing concentration measurement, concentration gradient value calculation, and maximum gradient direction determination, with the currently hovering unmanned aerial vehicle as a center and the other unmanned aerial vehicle as a flying unmanned aerial vehicle, and controlling the flying unmanned aerial vehicle to move to an intersection of the maximum gradient direction and a flight circle to hover.

[0033] The trace termination judgment module is configured to determine, after each alternation of the alternating execution module, whether a trace termination condition is met according to a concentration gradient threshold and an unmanned aerial vehicle position change setting value, and if the trace termination condition is met, to terminate the trace process, and if the trace termination condition is not met, to continue the alternation.

[0034] The embodiments of the present specification can at least achieve the following beneficial effects:

[0035] I. The trace efficiency is significantly improved, and the positioning time is greatly shortened

[0036] 1. Breakthrough single unmanned aerial vehicle efficiency bottleneck: Compared with the single unmanned aerial vehicle zigzag / grid blind search (which needs to cover a large area and is easy to detour), the present method directly approaches the pollution source along the "maximum gradient direction" through "double unmanned aerial vehicles alternating flight + concentration gradient directional tracking", without redundant paths and reducing invalid flight distance.

[0037] 2. Realize continuous gradient tracking: The double unmanned aerial vehicles adopt an alternating mode of "hovering-flying-moving", the previous unmanned aerial vehicle hovers to provide a reference concentration, the latter unmanned aerial vehicle flies to detect the gradient, the roles are exchanged after moving, there is no interruption in the whole detection process, and the interval waiting of the single unmanned aerial vehicle "measurement-moving-remasurement" is avoided, the iteration times are greatly reduced compared with the single unmanned aerial vehicle scheme, and the positioning time of the pollution source in a small area (such as 5km2) can be shortened.

[0038] 3. Full circumferential detection without direction omission: By "≥8 equidistant measurement points + interpolation algorithm to construct a full circumferential gradient curve", the maximum gradient direction within 0°-360° is directly locked, without the need for multiple adjustments of the detection direction, avoiding the problem of "missing the optimal path" in single-direction measurement, further reducing the time-consuming of tracing.

[0039] II. Low hardware and operation cost, reducing the threshold of technology landing

[0040] 1. Low hardware investment: only 2 drones are needed to achieve efficient tracing, compared with the existing multi-drone solution (usually more than 5 drones), the hardware procurement cost is reduced; and the sensors and positioning devices carried by the drones are conventional mature components, without the need for customized high-cost equipment.

[0041] 2. Simple coordination and operation: without complex swarm intelligence algorithms (such as large-scale coordination logic for multi-drone task allocation and conflict avoidance), coordination is achieved through "double-machine alternating fly-by + fixed radius control", the system software development difficulty is low, and the operation personnel do not need professional algorithm background, significantly reducing the training cost and later maintenance cost.

[0042] III. High accuracy and reliability of tracing, reducing misjudgment and deviation

[0043] 1. Controllable gradient calculation accuracy: through the design of "fixed preset radius (30-100 meters adjustable) + fly-by trajectory deviation ≤5%", the distance between the two drones is stable, combined with the gradient calculation logic of "concentration difference absolute value / fixed distance", the gradient error caused by fluctuation of distance is avoided, and the gradient calculation accuracy is improved compared with the "dynamic distance measurement scheme".

[0044] 2. Avoiding local optimal solution trap: single drone is prone to misjudgment as pollution source due to "local high concentration area", while this method is based on "hill climbing algorithm + full circumferential gradient comparison", always moving along the "global maximum gradient direction", and through "continuous iteration position change judgment" (convergence if the change is less than the set value for 3 times in a row), ensuring that the final positioning point is the core area of the pollution source, rather than a local interference area, with high accuracy of tracing.

[0045] 3. Double protection of termination condition: using "gradient threshold + position convergence" double condition judgment, it avoids missing pollution source due to "gradient sudden drop" (such as pollution source diffusion boundary), and prevents infinite iteration caused by "small deviation of drone", ensuring the reliability of the positioning result when the tracing is terminated.

[0046] IV. Wide application scenarios, breaking through the limitations of traditional monitoring

[0047] 1. Terrain and regional adaptability: Compared with fixed monitoring stations (only covering preset areas and unable to move), drones can flexibly enter complex / inconvenient areas such as mountainous areas, industrial parks, and remote villages; and by adjusting the "flying radius" (up to 80-100 meters when the terrain is flat, and down to 30-50 meters in complex terrain areas), the pollution diffusion characteristics of different terrains can be adapted without the need to redesign the scheme for different scenarios.

[0048] 2. High compatibility of pollution types: It can adapt to various air pollution scenarios such as industrial waste gas (such as VOCs, particulate matter), regional dust, and sudden leakage, and only needs to replace the corresponding type of air pollution sensor (such as PID sensor for detecting VOCs, and laser scattering sensor for detecting particulate matter) according to the type of pollution. The system core logic does not need to be changed, and the adaptability is strong.

[0049] Five, data value extension, supporting subsequent pollution control

[0050] 1. Full-process data traceability: Real-time recording of "drone position, concentration at each measurement point, gradient value" during the tracing process, and generation of "tracing path report + concentration gradient change curve" after the tracing is terminated, which clearly restores the pollution source positioning process and facilitates subsequent verification and review.

[0051] 2. Auxiliary pollution diffusion analysis: Based on the recorded concentration gradient data and location information, the pollution diffusion speed and diffusion range can be deduced (such as combined with the Gaussian diffusion model), providing data support for the environmental protection department to develop "pollution control range, emergency disposal scheme", realizing the closed loop of "positioning-control-control". BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, a brief introduction will be given below to the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0053] Figure 1 The flowchart of the pollution source rapid tracing method based on the double-drone cooperative gradient detection involved in some embodiments of the present application.

[0054] Figure 2 The schematic diagram of the double-drone cooperative working mode involved in some embodiments of the present application.

[0055] Figure 3 The schematic diagram of the circular flying measurement point distribution involved in some embodiments of the present application.

[0056] Figure 4Schematic diagram of software and hardware modules that can be used for the method involved in some embodiments of the present application.

[0057] Figure 5 Schematic diagram of the working effectiveness of the method involved in some embodiments of the present application under the simulation of Gaussian plume pollution concentration distribution.

[0058] Figure 6 Schematic diagram of the working effectiveness of the concentric circle encryption observation method involved in some embodiments of the present application under the actual pollution concentration distribution.

[0059] Figure 7 Schematic diagram of the working effectiveness of the traditional encryption observation method involved in some embodiments of the present application under the actual pollution concentration distribution (efficiency is 50% lower). DETAILED DESCRIPTION

[0060] Hereinafter, only certain exemplary embodiments are simply described. As can be appreciated by those skilled in the art, the described embodiments can be modified in various different manners without departing from the spirit or scope of the present application. Therefore, the drawings and the description are considered to be essentially exemplary rather than limiting.

[0061] The terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0062] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0063] As shown in the drawings, the present embodiment provides a pollution source rapid tracing method based on double unmanned aerial vehicle cooperative gradient detection, which comprises: Figure 1 S101: initialize system parameters, the system parameters including a preset radius of unmanned aerial vehicle flying (i.e. a fixed distance between two unmanned aerial vehicles), an equal angle interval point number when the unmanned aerial vehicle flies, a concentration gradient threshold value for determining the termination of tracing, and an unmanned aerial vehicle position change setting value for determining the convergence of tracing;

[0064] S102: deploy the first unmanned aerial vehicle to hover at the initial position, measure the atmospheric pollutant concentration value of the initial position by the atmospheric pollution sensor carried by the first unmanned aerial vehicle, and record it as the first concentration value;

[0065] S103: the first unmanned aerial vehicle flies to the second unmanned aerial vehicle at a preset radius, and the second unmanned aerial vehicle flies to the first unmanned aerial vehicle at a preset radius, and the two unmanned aerial vehicles fly to the next equal angle interval point at a preset radius;

[0066] S103: control the second unmanned aerial vehicle to fly around the first unmanned aerial vehicle as the center and the preset radius as the circumference, stop at a plurality of equiangular interval points respectively, measure the atmospheric pollutant concentration values of each stop point by the atmospheric pollution sensor carried by the second unmanned aerial vehicle, and record the second concentration values;

[0067] S104: according to the first concentration values, the second concentration values and the preset radius, calculate the atmospheric pollutant concentration gradient values of the second unmanned aerial vehicle at each equiangular interval point and the first unmanned aerial vehicle;

[0068] S105: based on the concentration gradient values calculated in S104, determine the direction with the maximum atmospheric pollutant concentration gradient in the full range of 0°-360°, and record the maximum gradient direction;

[0069] S106: control the second unmanned aerial vehicle to move to the intersection position of the maximum gradient direction and the fly-around circumference in S103, and hover at the intersection position;

[0070] S107: control the first unmanned aerial vehicle to fly around the second unmanned aerial vehicle hovering in S106 as the center and the preset radius as the circumference, stop at a plurality of equiangular interval points respectively and measure the atmospheric pollutant concentration values of each stop point, and repeat the concentration gradient measurement and maximum gradient direction determination process of S104-S105;

[0071] S108: alternately perform the operations of S103-S107, that is, sequentially perform the concentration measurement, concentration gradient value calculation and maximum gradient direction determination by taking the currently hovering unmanned aerial vehicle as the center and the other unmanned aerial vehicle as the fly-around unmanned aerial vehicle, and control the fly-around unmanned aerial vehicle to move to the intersection of the maximum gradient direction and the fly-around circumference to hover;

[0072] S109: after each alternately performing S103-S108, determine whether the tracing termination condition is met according to the concentration gradient threshold value and the unmanned aerial vehicle position change setting value, if the tracing termination condition is met, terminate the tracing process, if the tracing termination condition is not met, continue to alternately perform.

[0073] In some embodiments, the preset radius of the unmanned aerial vehicle fly-around is in the range of 30 meters to 100 meters, which is determined according to the terrain complexity of the monitoring area and the atmospheric pollutant diffusion speed. The flatter the terrain and the faster the diffusion speed, the larger the preset radius value.

[0074] In some embodiments, the number of equiangular interval points when the second unmanned aerial vehicle flies around is at least 8, and the angle difference between each equiangular interval point is equal. By increasing the number of equiangular interval points, the accuracy of concentration gradient detection in the full range of 0°-360° is improved.

[0075] In some embodiments, when calculating the concentration gradient value, the absolute value difference between the second concentration value and the first concentration value at each equiangular interval point is calculated based on the first concentration value of the first UAV, and then the absolute value difference is divided by the preset radius to obtain the concentration gradient value corresponding to the equiangular interval point.

[0076] In some embodiments, the specific process of determining the maximum gradient direction is as follows: first, the interpolation algorithm is used to process the concentration gradient values of the obtained equiangular interval points to construct a complete 0°-360° atmospheric pollutant concentration gradient distribution curve, and then the gradient distribution curve is analyzed to find the extreme point in the curve, and the direction corresponding to the extreme point is the maximum gradient direction in the 0°-360° full range.

[0077] In some embodiments, in S107, the first UAV flies around the second UAV as the center, and the flying operation of the second UAV in S103 needs to be consistent, that is, the same preset radius and the number of equiangular interval points are used, and the atmospheric pollutant concentration values are measured at each equiangular interval point, to ensure the uniformity of the parameters of the two flying measurements to improve the gradient calculation accuracy.

[0078] In some embodiments, in S108, when the operations are alternately performed, after the UAV is moved to the intersection position and hovers, the hovering position needs to be confirmed for accuracy by the positioning device carried by the UAV. If the deviation between the hovering position and the intersection position exceeds the preset accuracy value (which is a part of the system parameters), the UAV position is adjusted to meet the accuracy requirement before the next flying operation is performed.

[0079] In some embodiments, the specific way of judging the termination condition of the tracing in S109 is as follows: if the currently calculated concentration gradient value is lower than the concentration gradient threshold value, it is determined that the vicinity of the pollution source has been reached; if the UAV position changes in the last 3 iterations are all less than the UAV position change setting value, it is determined that the tracing process converges; and any of the above conditions is met to terminate the tracing.

[0080] In some embodiments, in the UAV flying process in S103 and S107, the flight speed needs to be kept stable, and the deviation between the flying trajectory and the circle with the hovering UAV as the center and the preset radius as the radius is not more than 5%, to avoid the change of the distance between the two UAVs due to the deviation of the trajectory, which affects the accuracy of the concentration gradient calculation.

[0081] The pollution source rapid tracing system based on the double-UAV collaborative gradient detection applies the pollution source rapid tracing method based on the double-UAV collaborative gradient detection in any of the above embodiments, and the pollution source rapid tracing system based on the double-UAV collaborative gradient detection comprises:

[0082] An initialization module is configured to initialize system parameters, including a preset radius of the UAV fly-around, a number of equi-angle interval points during the UAV fly-around, a concentration gradient threshold for determining the termination of the source tracing, and a UAV position change setting value for determining the convergence of the source tracing.

[0083] A first measurement module is configured to control the first UAV to hover at an initial position, and measure an atmospheric pollutant concentration value at the initial position by using an atmospheric pollutant sensor carried by the first UAV, which is recorded as a first concentration value.

[0084] A second measurement module is configured to control the second UAV to fly around the first UAV at a preset radius, and stop at a plurality of equi-angle interval points, and measure atmospheric pollutant concentration values at the stopping points by using an atmospheric pollutant sensor carried by the second UAV, which are recorded as second concentration values.

[0085] A first calculation module is configured to calculate atmospheric pollutant concentration gradient values of the second UAV at the equi-angle interval points relative to the first UAV according to the first concentration value, the second concentration values, and the preset radius.

[0086] A second calculation module is configured to determine a direction with the largest atmospheric pollutant concentration gradient within a 0°-360° full range based on the concentration gradient values calculated by the first calculation module, which is recorded as a maximum gradient direction.

[0087] A UAV moving module is configured to control the second UAV to move to an intersection position of the maximum gradient direction and a fly-around circle in the second measurement module, and hover at the intersection position.

[0088] A direction determination module is configured to control the first UAV to fly around the second UAV hovering in the UAV moving module as a center at a preset radius, and stop at a plurality of equi-angle interval points to measure atmospheric pollutant concentration values at the stopping points, and repeat the concentration gradient measurement and the maximum gradient direction determination.

[0089] An alternating execution module is configured to alternately execute operations of: sequentially performing concentration measurement, concentration gradient value calculation, and maximum gradient direction determination by taking a currently hovering UAV as a center and another UAV as a fly-around UAV, and controlling the fly-around UAV to move to an intersection position of the maximum gradient direction and a fly-around circle to hover.

[0090] A source tracing termination judgment module is configured to determine whether a source tracing termination condition is met according to the concentration gradient threshold and the UAV position change setting value after each alternation of the alternating execution module, and terminate the source tracing process if the source tracing termination condition is met, or continue the alternation if the source tracing termination condition is not met.

[0091] The technical concept of the application is as follows:

[0092] S101: Initialize system parameters, including the UAV's flight radius R (usually set to 30-100 meters), the number of measurement points N (usually no less than 8), gradient threshold ε, etc.

[0093] S102: Deploy the first UAV to hover at the initially estimated location or a randomly selected location and measure the pollutant concentration value C1 at that location;

[0094] S103: Control the second drone to fly in a circle around the first drone with radius R as the center, and measure the pollutant concentration C at N equally spaced angular points. 21 C 22 ,...,C 2n ;

[0095] S104: Calculate the concentration gradient value between each measurement point and the first UAV (Gradient) Gᵢ=|C2ᵢ-C1| / R;

[0096] S105: By comparing the gradient values ​​in each direction, determine the direction with the largest pollutant concentration gradient within the entire circumference range of 0° to 360°. ;

[0097] S106: Control the second drone to move to the direction of maximum gradient. The point of intersection with the circumference of the flight path and hovering;

[0098] S107: Control the first UAV to fly in a circle around the second UAV with radius R as the center, repeating the above process of concentration gradient measurement and determination of maximum gradient direction.

[0099] S108: Alternately execute the above steps (to achieve rapid source tracing of pollution based on hill climbing algorithm) until the termination condition is met;

[0100] S109: When the concentration gradient value is lower than the preset threshold ε or the change in the drone's position is less than the set value in multiple consecutive iterations, it is determined that the area near the pollution source has been reached, and the source tracing process is terminated.

[0101] like Figure 2 and Figure 3 The diagram illustrates the collaborative operation of two drones. Initially, drone 1 hovers at point A, while drone 2 flies around it to measure and determine the direction of the maximum gradient as α, then moves to point B and hovers. Next, drone 1 flies around drone 2, determines the direction of the maximum gradient as β, and moves to point C. This process is repeated alternately, gradually approaching the pollution source S. The collaborative source tracing trajectory of the two drones is shown below. Figure 5 As shown.

[0102] In some embodiments:

[0103] I. System Initialization (S101)

[0104] Initialize key system parameters, including:

[0105] The reference radius R of the fly-around (usually 30-100 meters, which can be dynamically adjusted according to the complexity of the terrain and the speed of pollution diffusion);

[0106] The number of equally spaced measurement points N during the fly-around (initially set to no less than 8 to ensure the basic accuracy of the full-circumferential gradient detection);

[0107] The trace termination gradient threshold = 0.01 (physically, it means the concentration gradient, i.e., the concentration change per unit distance);

[0108] The position convergence judgment threshold (30 meters, used to determine whether the UAV position change is tending to be stable).

[0109] II. First concentration measurement and gradient analysis (S102-S105)

[0110] 1. S102: Reference point concentration measurement

[0111] Deploy the first UAV to hover at the initial estimated position (or randomly selected potential pollution area) and measure the pollutant concentration value at this position as the first concentration reference point for the entire process.

[0112] 2. S103: Circumferential fly-around concentration collection

[0113] Control the second UAV to fly around the first UAV as the center with R as the radius, hover at N equally spaced points (0°, 45°, 90°, etc., with an interval angle of 360° / N), and measure the pollutant concentration value at each point to obtain the full-circumferential concentration distribution data.

[0114] 3. S104: Gradient value calculation

[0115] Calculate the concentration gradient value of each interval point with the reference point (the first UAV) according to the formula , which quantifies the concentration change rate in each direction.

[0116] 4. S105: Maximum gradient direction determination

[0117] Compare all gradient values , determine the direction with the maximum concentration gradient (denoted as ) within the 0°-360° full-circle range, and use it as the initial trace tracking direction.

[0118] III. Special scene processing after the first fly-around (S106) ​​​

[0119] According to the results of S103-S105, the following special processing flow is triggered:

[0120] Scenario 1: The gradient difference in each direction is extremely small

[0121] If the difference between the gradient values of all interval points is ≤ (0.01 ug / m 4 ), it indicates that the initial N measurement points are not accurate enough, and the measurement point encryption process is started: the number of equiangular interval points around the fly is encrypted from N to 36 (interval 10°), and S103-S105 are re-executed to avoid misjudgment of the gradient direction due to sparse measurement points.

[0122] Scenario 2: Multiple local high-value regions coexist

[0123] If one gradient large value direction is initially detected, but after encryption (36 points) it is found that there are two or more gradient large value regions, and the angle between the regions is > 10°, it indicates that there are multiple pollution sources or pollution diffusion multi-directionality, and the double-machine concentric circle encryption monitoring (S111-S115) is started to capture all local high-value regions; after completion, the region with the largest gradient is selected to restart the hill climbing algorithm (S107-S110).

[0124] Scenario 3: No special case

[0125] If the gradient direction is clear (single maximum gradient direction), directly enter the hill climbing algorithm flow (S107-S110).

[0126] Four, hill climbing algorithm tracing (S107-S110)

[0127] 1. S107: Move to the point with the largest gradient

[0128] Control the second unmanned aerial vehicle to move to the "maximum gradient direction Intersection with the fly-around circle" position and hover, as a new concentration measurement reference point.

[0129] 2. S108: Role switching and gradient re-measurement

[0130] Control the first unmanned aerial vehicle to fly around the "hovering second unmanned aerial vehicle" as the center and R as the radius, repeat S103 (measure the concentration of interval points) -S104 (calculate the gradient) -S105 (determine the new maximum gradient direction).

[0131] Note: If S106 has started 36-point encryption, this step and subsequent fly-around use 36 measurement points to ensure accuracy.

[0132] 3. S109: Alternating iterative approximation

[0133] Alternately execute S107-S108 (i.e. continue switching between the "circle center UAV" and "fly around UAV" roles), gradually move in the direction of the maximum gradient determined each time, and achieve "ladder approximation of pollution sources".

[0134] 4. S110: Trace termination determination

[0135] Check the termination condition after each iteration. If any of the following conditions are met, terminate the trace and determine that the pollution source has been reached:

[0136] Current gradient value < 0.01 ug / m3 (0.01 ug / m3 4 ), indicating that the pollution source core has been approached and the concentration change is tending to be flat;

[0137] In the last 3 iterations, the UAV position change is ≤30 meters (position convergence, no obvious direction adjustment space).

[0138] Five, concentric circle encryption monitoring (S111-S115)

[0139] Only in S106 "Scenario 2" triggers, used to investigate multiple local high value area:

[0140] 1. S111: Determine the concentric circle reference point

[0141] Record the "current hovering position of the first UAV" after completing the encryption observation in S106, denoted as circle center , to ensure that the concentric circle origin is consistent with the latest measurement reference.

[0142] 2. S112: Close-range fly-around investigation

[0143] Control the second UAV to fly around (36 equidistant points) with as the center and 50 meters as the radius, measure the concentration at each point, and investigate local high values in close range.

[0144] 3. S113: Medium-distance fly-around investigation

[0145] Control the first UAV to fly around (36 equidistant points) with as the center and 100 meters as the radius, expanding the investigation range to the medium-distance area.

[0146] 4. S114: Sub-long-distance fly-around investigation

[0147] Control the second UAV to fly around (36 equidistant points) with as the center and 150 meters as the radius, continuously expanding outward; if a point gradient value < 0.01 ug / m3 is detected during the fly-around, the fly-around at this radius can be terminated in advance, and the climbing algorithm for this area is started preferentially.

[0148] 5. S115: Long-distance fly-around and termination

[0149] The first unmanned aerial vehicle is controlled to fly around as the center and 200 meters as the radius (36 equiangular points) until the unmanned aerial vehicle is insufficient in power; after the end, based on all the high-value areas found, a hill climbing algorithm is started in turn according to the gradient from large to small.

[0150] Six, process connection explanation

[0151] 1. The measurement point encryption (36 points) is continuously effective after triggering until the trace termination or switching to the concentric circle mode;

[0152] 2. The concentric circle encryption monitoring is an "interrupted process", and after completion, the hill climbing algorithm is returned to track the high-value areas found one by one;

[0153] 3. In all fly-around operations, if the power is lower than the safety threshold (such as the remaining power only supports returning to the starting point), the current task is terminated and returned in priority, to ensure the safety of the equipment.

[0154] The working effect of the concentric circle encryption observation method under the actual pollution concentration distribution is as shown in Figure 6 The working effect of the traditional encryption observation method under the actual pollution concentration distribution is as shown in Figure 7

[0155] The above method can be implemented by using the following software and hardware modules, as shown in Figure 4 , including a first unmanned aerial vehicle unit 401, a second unmanned aerial vehicle unit 402, a ground control station 403, a gradient calculation module 404, a direction decision module 405, and a trace termination judgment module 406. The ground control station 403 further includes a path planning module and a result visualization module.

[0156] The first unmanned aerial vehicle unit 401 and the second unmanned aerial vehicle unit 402 are both equipped with high-precision atmospheric pollution sensors, GPS positioning devices, wireless communication modules, and flight control systems; the ground control station 403 is responsible for overall task coordination and data fusion processing; the gradient calculation module 404 is responsible for processing the concentration measurement data of the two unmanned aerial vehicles and calculating the concentration gradient; the direction decision module 405 is responsible for determining the maximum gradient direction and generating unmanned aerial vehicle movement instructions; the trace termination judgment module 406 is responsible for determining whether to terminate the trace process according to the preset conditions; the path planning module is responsible for generating the fly-around path and movement path of the unmanned aerial vehicle; and the result visualization module is responsible for real-time display of the unmanned aerial vehicle position, pollution concentration distribution, and trace path.

[0157] The present application realizes efficient and low-cost pollution source rapid tracing through the cooperative working mode of the double unmanned aerial vehicles, and provides a new technical means for the environmental monitoring field.

[0158] ​​In summary, the multiple specific embodiments of the present application are disclosed, each of the embodiments can be freely combined to form new embodiments without being self-contradictory, that is, the embodiments belonging to the alternative solutions can be freely replaced, but cannot be combined with each other; the embodiments not belonging to the alternative solutions can be combined with each other, and these new embodiments also belong to the substantial content of the present application.

[0159] The above embodiments describe multiple specific implementations of the present application, but those skilled in the art should understand that various changes or modifications can be made to these implementations without departing from the principles and essence of the present application, and such changes and modifications all fall within the protection scope of the present application.

Claims

1. A pollution source fast tracing method based on double-UAV cooperative gradient detection, characterized in that, The method comprises the following steps: S101: initializing system parameters, the system parameters comprising a preset radius of the UAV fly-around, a number of equiangular interval points during the UAV fly-around, a concentration gradient threshold value for determining the termination of the tracing, and a UAV position change setting value for determining the convergence of the tracing; S102: deploying a first UAV to hover at an initial position, and measuring an atmospheric pollutant concentration value at the initial position by an atmospheric pollutant sensor carried by the first UAV, the value being recorded as a first concentration value; S103: controlling a second UAV to fly around the first UAV with the preset radius as the circumference, and stopping at a plurality of equiangular interval points, and measuring atmospheric pollutant concentration values at the stopping points by an atmospheric pollutant sensor carried by the second UAV, the values being recorded as second concentration values; S104: calculating atmospheric pollutant concentration gradient values of the second UAV at the equiangular interval points relative to the first UAV according to the first concentration value, the second concentration values, and the preset radius; S105: determining a direction with the largest atmospheric pollutant concentration gradient within a full range of 0°-360° based on the concentration gradient values calculated in S104, and recording the direction as a maximum gradient direction; S106: controlling the second UAV to move to an intersection position of the maximum gradient direction and the fly-around circumference in S103, and hovering at the intersection position; S107: controlling the first UAV to fly around the second UAV hovering in S106 with the preset radius as the circumference, stopping at a plurality of equiangular interval points, measuring atmospheric pollutant concentration values at the stopping points, and repeating the concentration gradient measurement and maximum gradient direction determination processes in S104-S105; S108: alternately performing the operations in S103-S107, i.e., sequentially performing concentration measurement, concentration gradient value calculation, and maximum gradient direction determination with the currently hovering UAV as the center and the other UAV as the fly-around UAV, and controlling the fly-around UAV to move to the intersection of the maximum gradient direction and the fly-around circumference to hover; S109: after each alternately performing S103-S108, determining whether a tracing termination condition is met according to the concentration gradient threshold value and the UAV position change setting value, and if the condition is met, terminating the tracing process, and if the condition is not met, continuing the alternately performing.

2. The pollution source fast tracing method based on dual-UAV cooperative gradient detection according to claim 1, characterized in that, The preset radius of the UAV fly-around ranges from 30 meters to 100 meters, and the value is determined according to the terrain complexity of the monitoring area and the atmospheric pollutant diffusion speed. The flatter the terrain and the faster the diffusion speed, the larger the preset radius value.

3. The method of claim 1, wherein, The number of equiangular interval points during the fly-around of the second UAV is at least 8, and the angle difference between each equiangular interval point is equal. The accuracy of the concentration gradient detection within the full range of 0°-360° is improved by increasing the number of equiangular interval points.

4. The method of claim 1, wherein, When calculating the concentration gradient value, the first concentration value of the first UAV is taken as the reference, the absolute value difference between the second concentration value and the first concentration value at each equiangular interval point is calculated, and then the absolute value difference is divided by the preset radius to obtain the concentration gradient value of the corresponding equiangular interval point.

5. The method of claim 1, wherein, The specific process of determining the maximum gradient direction is: first, the interpolation algorithm is used to process the concentration gradient values of the obtained points at equal angle intervals, to construct a complete 0°-360° atmospheric pollutant concentration gradient distribution curve, and then the gradient distribution curve is analyzed to find the extreme point in the curve, and the direction corresponding to the extreme point is the maximum gradient direction in the full 0°-360° range.

6. The method of claim 1, wherein, In S107, the first unmanned aerial vehicle flies around the second unmanned aerial vehicle as the center, and the flying operation of the second unmanned aerial vehicle in S103 needs to be consistent, that is, the same preset radius and the same number of equal angle interval points are used, and the atmospheric pollutant concentration values are measured at each equal angle interval point, to ensure the uniformity of the parameters of the two flying measurements to improve the gradient calculation accuracy.

7. The method of claim 1, wherein, In S108, when the operations are alternately performed, after the unmanned aerial vehicle is moved to the intersection position and hovers, the accuracy of the hovering position needs to be confirmed by the positioning device carried by the unmanned aerial vehicle. If the deviation between the hovering position and the intersection position exceeds the preset accuracy value, the position of the unmanned aerial vehicle is adjusted to meet the accuracy requirement before the next flying operation is performed.

8. The method of claim 1, wherein, In S109, the specific way to determine the termination condition of the tracing is: if the currently calculated concentration gradient value is lower than the concentration gradient threshold value, it is determined that the vicinity of the pollution source has been reached; if the position change of the unmanned aerial vehicle in the last 3 iterations is less than the set value of the position change of the unmanned aerial vehicle, it is determined that the tracing process converges; if any of the above conditions is met, the tracing is terminated.

9. The method of claim 3, wherein, In S106, according to the measurement and calculation results of S103-S105, the corresponding special processing procedure is triggered: If the difference between the concentration gradient values of all equiangular interval points is less than or equal to 0.01 ug / m 4 If the difference between the concentration gradient values of all equiangular interval points is less than or equal to 0.01 ug / m 4 , the measurement point encryption process is started, the number of fly-around equiangular interval points is encrypted from the initial at least 8 to 36, the angle difference between each interval point is 10°, and S103-S105 are re-executed. If one concentration gradient maximum value direction is initially detected, and after encryption to 36 equal angle interval points, it is found that there are 2 or more gradient maximum value regions, and the included angle between each gradient maximum value region is greater than 10°, then the double-machine concentric circle encryption monitoring process is started, which includes: S111: records the current hovering position of the first unmanned aerial vehicle after the completion of the encryption observation in S106, which is recorded as the center of the concentric circle; S112: controls the second unmanned aerial vehicle to fly around the center with a radius of 50 meters, and measures the atmospheric pollutant concentration values at 36 equal angle interval points; S113: controls the first unmanned aerial vehicle to fly around the center with a radius of 100 meters, and measures the atmospheric pollutant concentration values at 36 equal angle interval points; S114: control the second unmanned aerial vehicle to fly around the center of the circle with a radius of 150 meters, measure the atmospheric pollutant concentration value at each point on the 36 equiangularly spaced points, and if the concentration gradient value of a point is detected to be <0.01 ug / m 4 during the flying around, terminate the flying around operation of the radius in advance, and restart the operation of S107-S109 in priority to the area where the point is located; S115: controls the first unmanned aerial vehicle to fly around the center with a radius of 200 meters, and measures the atmospheric pollutant concentration values at 36 equal angle interval points until the remaining power of the unmanned aerial vehicle is lower than the safety threshold; after the completion of the double-machine concentric circle encryption monitoring process, based on all the gradient maximum value regions found, the operation of S107-S109 is restarted in the order of the concentration gradient value from large to small; In S107-S109, if the measurement point encryption process has been started in S106, that is, the encryption is performed to 36 equal angle interval points, then the flying operations of the first unmanned aerial vehicle and the second unmanned aerial vehicle are both based on the 36 equal angle interval points.

10. A pollution source rapid tracing system based on dual-UAV cooperative gradient detection, characterized in that, The double-unmanned aerial vehicle cooperative gradient detection based pollution source rapid tracing method according to any one of claims 1-9, the double-unmanned aerial vehicle cooperative gradient detection based pollution source rapid tracing system comprises: The initialization module is configured to initialize system parameters, including a preset radius of the UAV fly-around, a number of equiangular interval points during the UAV fly-around, a concentration gradient threshold for determining termination of the source tracing, and a UAV position change setting value for determining convergence of the source tracing; The first measurement module is configured to control the first UAV to hover at an initial position, and measure an atmospheric pollutant concentration value at the initial position by using an atmospheric pollutant sensor carried by the first UAV, which is recorded as a first concentration value; The second measurement module is configured to control the second UAV to fly around the first UAV at a preset radius, and stop at a plurality of equiangular interval points, and measure atmospheric pollutant concentration values at the stopping points by using an atmospheric pollutant sensor carried by the second UAV, which are recorded as second concentration values; The first calculation module is configured to calculate atmospheric pollutant concentration gradient values of the second UAV at the equiangular interval points relative to the first UAV according to the first concentration value, the second concentration values, and the preset radius; The second calculation module is configured to determine a direction with the largest atmospheric pollutant concentration gradient within a full range of 0°-360° based on the concentration gradient values calculated by the first calculation module, which is recorded as a maximum gradient direction; The UAV moving module is configured to control the second UAV to move to an intersection position of the maximum gradient direction and a fly-around circle, and hover at the intersection position; The direction determination module is configured to control the first UAV to fly around the second UAV hovering at the intersection position as a center and the preset radius as a circle, stop at a plurality of equiangular interval points, and measure atmospheric pollutant concentration values at the stopping points, and repeat the concentration gradient measurement and the maximum gradient direction determination; The alternate execution module is configured to alternately execute operations of: sequentially taking a currently hovering UAV as a center and another UAV as a fly-around UAV to measure concentrations, calculate concentration gradient values, and determine a maximum gradient direction, and control the fly-around UAV to move to an intersection position of the maximum gradient direction and a fly-around circle to hover; The source tracing termination judgment module is configured to determine whether a source tracing termination condition is met according to the concentration gradient threshold and the UAV position change setting value after each alternation of the alternate execution module, and if the source tracing termination condition is met, terminate the source tracing process, and if the source tracing termination condition is not met, continue the alternation.

Images