Multi-machine collaborative forest fire monitoring optimal deployment method

Abstract

The invention discloses a multi-aircraft cooperative forest fire monitoring optimal deployment method, and belongs to the technical field of unmanned aerial vehicle emergency rescue. The method comprises the following steps: firstly, receiving a fire report, estimating a fire position according to the fire report, and establishing a forest fire spreading grid model; then the fire position is sent to the UAV, and a UAV cluster is started to carry out forest fire reconnaissance; and setting a target function according to the importance degree of each block in the live wire and the fire scene image quality in the investigation process, and determining the optimal coverage deployment of the UAV cluster for the fire scene according to the target function. According to the method, the UAV pose can be dynamically regulated and controlled according to the fire scene spreading situation, the safety constraints such as multi-aircraft collision prevention and burning prevention are met, the high-precision perception of the fire spreading front edge can be ensured while the fire wire coverage rate is maximized, and the method has environmental robustness for coping with the disturbance of multiple factors such as the wind direction and the terrain.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) emergency rescue technology, specifically involving an optimal deployment method for multi-aircraft collaborative forest fire monitoring. Background Technology

[0002] Forest fires are extremely destructive, often causing severe losses and chaos in a short period. Fighting forest fires is dangerous and urgent; failure to promptly grasp the dynamic changes and real-time situation of the fire can easily lead to accidents. If the fire situation changes abruptly and spreads suddenly to the area where rescue personnel are located, firefighters will face extremely high risks to their lives. Therefore, it is necessary to accurately determine the location of the forest fire, and more importantly, to accurately observe its development to track its spread boundaries. The more information gathered about the area of ​​fire spread, the better the on-site commander can develop plans to evacuate people and property from danger zones and effectively prevent the fire from spreading to new areas.

[0003] Using UAVs to assist in fighting forest fires and other natural disaster relief efforts is very promising. They can assist humans in fire-tracking missions and replace manned helicopters, saving considerable operating costs compared to traditional methods. However, research discussing the application of UAVs in firefighting remains limited.

[0004] Specifically, when using UAV clusters to track and reconnoiter the spread of forest fires, real-time adjustment of the UAVs' positions and orientations is necessary. This includes avoiding collisions between UAVs and ensuring a safe distance between them and the ground to prevent them from being destroyed by flames. Furthermore, the UAV cluster needs to balance two performance metrics: fire front coverage and fire field perception accuracy, to achieve optimal deployment for forest fire monitoring. Moreover, considering the uncertainty inherent in forest fire spread due to factors such as terrain, wind speed and direction, forest combustible load, and temperature, multi-UAV deployment schemes need to adapt to dynamic environmental changes and capture the trends of fire spread boundaries to maintain efficient reconnaissance of the fire field. However, currently, there is no multi-UAV collaborative forest monitoring deployment scheme that comprehensively considers all these factors. Summary of the Invention

[0005] In view of this, based on safety constraints such as multi-machine collision avoidance and fire prevention, the requirements of forest fire reconnaissance mission effectiveness (fire line coverage and fire scene image perception accuracy) and dynamic environmental adaptability, this invention proposes an optimal deployment method for multi-machine collaborative forest fire monitoring, which is used to monitor the spread of forest fires in real time.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides an optimal deployment method for multi-machine collaborative forest fire monitoring, comprising:

[0008] Receive fire reports;

[0009] The location of the fire is estimated based on the fire report, and a forest fire spread grid model is established. The attributes of each grid cell in the forest fire spread grid model include combustible material type, terrain elevation, and combustion state. The combustion state includes unburned, incompletely burned, completely burned, and burnt out.

[0010] The location of the fire is sent to the UAV, and the UAV cluster is activated to conduct forest fire reconnaissance.

[0011] An objective function is set based on the importance of each block in the fire line and the quality of the fire scene images. The optimal coverage deployment of the UAV cluster for the fire scene is determined based on the objective function. The fire line is composed of grid units that are in a state of incomplete combustion, and each grid unit corresponds to a block in the fire line. The fire scene images are captured by cameras equipped with UAVs.

[0012] Secondly, the present invention provides an optimal deployment system for multi-machine collaborative forest fire monitoring, comprising:

[0013] The receiving module is used to receive fire reports;

[0014] The location estimation and model building module is used to estimate the fire location based on the fire report and build a forest fire spread grid model. The attributes of each grid cell in the forest fire spread grid model include combustible material type, terrain height, and combustion state. The combustion state includes unburned, incompletely burned, completely burned, and burnt out.

[0015] The startup module is used to send the fire location to the UAV and start the UAV cluster for forest fire reconnaissance;

[0016] The deployment module is used to set an objective function based on the importance of each block in the fire line and the quality of the fire scene images, and to determine the optimal coverage deployment of the UAV cluster for the fire scene based on the objective function; the fire line is composed of grid units in the incomplete combustion state, and each grid unit corresponds to a block in the fire line; the fire scene images are captured by cameras equipped with UAVs.

[0017] Thirdly, the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors enable the aforementioned multi-machine collaborative optimal deployment method for forest fire monitoring.

[0018] Fourthly, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enable the processor to implement the aforementioned optimal deployment method for multi-machine collaborative forest fire monitoring.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention can dynamically adjust the UAV's position and attitude according to the fire spread situation, which not only meets the safety constraints such as multi-aircraft collision avoidance and fire prevention, but also ensures high-precision perception of the fire spread front while maximizing fire line coverage, and has environmental robustness to cope with disturbances from multiple factors such as wind direction and terrain. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the optimal deployment method for multi-machine collaborative forest fire monitoring according to the present invention.

[0022] Figure 2 This is a schematic diagram of the forest fire spread grid model established in this invention.

[0023] Figure 3 This is a schematic diagram of the UAV's field of view.

[0024] Figure 4 This is a schematic diagram of an optimal deployment method for multi-machine collaborative forest fire monitoring according to an embodiment of the present invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0027] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0028] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0029] Furthermore, to better illustrate the present invention, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that the present invention can be practiced without certain specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order to highlight the spirit of the invention.

[0030] Figure 1 A flowchart illustrating the optimal deployment method for multi-machine collaborative forest fire monitoring according to the present invention is shown, such as... Figure 1 As shown, the method may include:

[0031] S101, Receive fire report.

[0032] S102. Estimate the location of the fire based on the fire report and establish a forest fire spread grid model; the attributes of each grid cell in the forest fire spread grid model include combustible material type, terrain height, and combustion state; the combustion state includes unburned, incompletely burned, completely burned, and burned out.

[0033] Figure 2 A schematic diagram of the forest fire spread grid model established by this invention is shown, as follows: Figure 2 As shown, each grid cell integrates attributes such as combustible material type, terrain height, and combustion state. The set of grid cells in the incomplete combustion state is called the fire line, and each grid cell corresponds to a block within the fire line. The Line of Fire express.

[0034] S103. Send the fire location to the UAV and start the UAV cluster to conduct forest fire reconnaissance.

[0035] For example, the present invention employs an X-type quadcopter UAV, each equipped with positioning devices such as a Global Positioning System (GPS) and an Inertial Measurement Unit (IMU), as well as a camera capable of detecting fires, with the camera lens positioned vertically downwards. Each camera has a circular field of view, and the camera and its field of view form a cone with a half-angle of α. Figure 3 Showing the first in the UAV cluster UAV (i.e., drone) A schematic diagram of the field of view, wherein, Indicates drone Location coordinates, Indicates drone The set of all points within the field of view. The UAV field of view is the field of view of the camera equipped on the UAV.

[0036] Furthermore, assuming the communication radius of the UAV is If the distance between two UAVs is less than the communication radius Then the two UAVs can exchange information.

[0037] For example, UAVs can be deployed in warehouses around forests or at forest fire department centers. Once a fire report is received, the UAV activates and moves to the initially estimated location of the fire. The UAV will continue moving towards the fire location until it can detect the forest fire within its field of view.

[0038] S104. Set an objective function based on the importance of each block in the fire line and the quality of the fire scene images, and determine the optimal coverage deployment of the UAV cluster for the fire scene based on the objective function; the fire scene images are captured by cameras equipped with UAVs.

[0039] The following describes the process by which this invention determines the optimal coverage deployment of a UAV cluster for a fire scene.

[0040] To avoid collisions between UAVs, this invention designs a virtual repulsive force; if the distance between UAVs is less than a preset distance... ( If the repulsion force is applied, then the repulsion force takes effect. Define the first [repulsion force] in the UAV cluster. The first UAV and the first Collision avoidance potential field between UAVs for:

[0041] (1)

[0042] in, Indicates the first The position vectors of each UAV Indicates the first Position vectors of each UAV; It is a normal coefficient; Represents the magnitude of a vector.

[0043] set up This indicates all data in the UAV cluster related to the first... The distance between the UAVs is less than The set of UAVs, then the first The inter-machine collision avoidance repulsion force experienced by each UAV It can be done The gradient of the potential field generated by all UAVs is obtained as follows:

[0044] (2)

[0045] in, This represents the gradient operator.

[0046] Similarly, to maintain a safe distance between the UAV and the ground to prevent it from being burned by flames, a safe distance potential field is defined. for:

[0047] (3)

[0048] in, Indicates the first The z-axis coordinates of each UAV; It is a normal coefficient; Indicates the safe altitude of the UAV.

[0049] By taking the gradient of formula (3), we can obtain the first... The ground repulsion force experienced by each UAV for:

[0050] (4)

[0051] in, It is a unit vector pointing vertically upwards.

[0052] In this invention, the effectiveness of UAV fire reconnaissance missions includes fire line coverage and image quality (i.e., fire scene image perception accuracy) when cameras cover the spread of the fire. In practice, the perception area of ​​each UAV node can be considered as the shooting area of ​​the installed camera; therefore, it is generally assumed that the coverage quality of all points within the detection range is the same. Since each camera has a limited number of pixels to capture, when the UAV's field of view is large, fire scene images will be provided at a lower resolution. By reducing the UAV's circular field of view, i.e., lowering the UAV's flight altitude, higher resolution fire scene images can be obtained. The quality of the fire scene image can be measured by a detection quality function, which is set accordingly. for:

[0053] (5)

[0054] in, This represents the horizontal coordinates (i.e., coordinates on the horizontal plane) of the center point of the first block in Fireline. The first block can be any block in Fireline. , These represent the minimum and maximum altitude limits for UAV flight, respectively. Indicates the first The quality of the fire scene images obtained by one UAV capturing images of the first block. The higher the value, the higher the quality of the captured fire scene image.

[0055] If point If formula (6) is satisfied, then point exist within (i.e. ):

[0056] (6)

[0057] in, Indicates the first The horizontal position vector of each UAV.

[0058] It can be seen that, The value depends on the first Flight altitude of the UAV , The value ranges from 0 to 1. The larger, The smaller the value, the worse the quality of the fire scene image.

[0059] To ensure the coverage of the UAV cluster for the fireline, this invention designs a cost function. This is used to represent the cost when the first block is covered by multiple UAV fields of view. The calculation formula is as follows:

[0060] (7)

[0061] in, This represents the set of UAVs whose field of view covers the first block, when the first block is not covered by any UAV's field of view. It is an empty set, therefore To avoid division by zero errors, set It is a very small constant.

[0062] The final objective function is as follows. :

[0063] (8)

[0064] in, is a sensitivity function, representing the importance of the first block.

[0065] The optimal coverage deployment of a UAV cluster for a fire scene is shown in Equation (9):

[0066] (9)

[0067] in, Indicates the first A UAV in Position vector at any given time; This indicates the number of UAVs in the UAV cluster.

[0068] It's easy to see that when a UAV cluster covers the fireline, if its height remains constant, the cost of covering the uncovered fireline points (i.e., the center point of the fireline block) is a very large number, namely 1 / Therefore, UAV clusters tend to be spread out to cover more firing points, thereby reducing... If the horizontal deployment of the UAV cluster remains unchanged, lowering the altitude of one of the UAVs will improve its coverage quality. There is a decreasing trend; however, at the same time, the field of view of UAVs decreases with decreasing altitude, resulting in a reduction in the coverage of the fire line by the UAV cluster, thus prompting... Increase. Therefore, the objective function designed in this invention, as shown in formula (8), effectively balances the two task performance indicators of UAV clusters: fire coverage and fire scene image perception accuracy.

[0069] In the objective function shown in formula (8), The dynamic deployment of UAV clusters also has a significant impact. This invention assumes that UAVs equipped with infrared cameras can locate fire front areas using image recognition technology. Furthermore, UAVs are also equipped with anemometers to measure wind direction. Generally, the direction of fire spread is the outward normal to its boundary. However, in forest environments, fire spread tends to occur along higher terrain and is influenced by the dynamic factor of wind direction. For fire front points with faster spread rates, i.e., the fire head area, due to their higher risk and more urgent response requirements, they should be assigned a higher level of importance (i.e., greater importance). This allows for priority coverage and reconnaissance of the area, enabling more timely deployment of UAV clusters for forest fire monitoring to adapt to dynamic environmental changes.

[0070] Based on this, the present invention designs a sensitivity function as shown in formula (10). Used to measure the importance of each block in Fireline:

[0071] (10)

[0072] in, , , It is a constant, and ; Indicates the fire line is at the point. The direction of the outward normal at point The angle between the topographic gradients at that location Indicates the fire line is at the point. The angle between the direction of the outer normal and the wind direction (e.g.) Figure 1 As shown); Indicates the importance of the first block. The larger the value, the higher the importance of the first block, and the higher the priority of the first block being covered and detected by UAVs. (Fireline at point...) The direction of the outward normal at a point can be obtained by constructing a signed distance field (SDF). The terrain gradient at a given location can be obtained by performing a 3D model of the terrain.

[0073] The closer the outward normal direction of a fire line point is to the terrain gradient or wind direction, the more likely it is to spread outwards more quickly. As can be seen from formula (10), The range of values ​​is a certain point of or The smaller it is, the more important it is. The larger the fire, the more priority it needs to be for UAV detection. This enables dynamic deployment of multi-UAV collaborative forest fire monitoring with dynamic environmental adaptability.

[0074] The coverage and tracking behavior of each UAV in the UAV cluster along the forest fire edge (i.e., the fire line) is subject to the objective function. The guidance is manifested in For the The position vector of each UAV The negative gradient is shown below:

[0075] (11)

[0076] in, It is a constant; Indicates the first The coverage and tracking behavior of a UAV over the fire line is determined by the objective function. The guidance.

[0077] Define the first in the UAV cluster The field of view of a UAV is The set of fire lines that are covered at all times is , The endpoint set is ,Right now:

[0078] (12)

[0079] Then according to Objective function The integral is divided into three point sets, and the objective function after the division is... as follows:

[0080] (13)

[0081] in, express and The difference set, express and The difference set, This represents the difference operation.

[0082] The last addend in formula (13) and the first The position vector of each UAV It is irrelevant, therefore it is to The gradient is 0.

[0083] The first addend in formula (13) and the second addend The horizontal position vector of each UAV It is irrelevant to it. The gradient of is 0, therefore we can obtain the first . The horizontal control component of the UAV for:

[0084] (14)

[0085] in, Point Just in the th When the detection range of a UAV is at the boundary, point The unit outward normal vector; Indicates in set Remove the first The set obtained from each UAV. Horizontal control component. Used to control the The movement of a UAV on the horizontal plane guides it to track fire points with high coverage sensitivity function values, while reducing the detection overlap with other UAVs.

[0086] The first one can also be obtained using formula (13). The height control component of each UAV for:

[0087] (15)

[0088] Height control component Used to control the The flight altitude of each UAV. The first addend in formula (15) is the vertical downward gravity, which is used to improve the quality of fire scene images; the second addend is the vertical upward gravity, which is used to obtain a larger circular field of view; the two achieve a balance between the two mission performances of fire scene image perception accuracy and fire line coverage.

[0089] According to , get , yes and The sum vector, i.e. .

[0090] In summary, it can be based on , , Get the first Desired deployment locations for each UAV:

[0091] (16)

[0092] in, Indicates the first A UAV in The expected deployment location vector at time; Indicates the first A UAV in The first derivative of the position vector at time t; Indicates the quality of the UAV (all UAVs in the UAV cluster have the same quality); .

[0093] The present invention adopts the planning algorithm shown in formula (16) to realize a method for optimal deployment of forest fire monitoring with multi-machine collaboration that is safe in avoiding collisions and preventing burning, takes into account the fire line coverage and the accuracy of fire scene image perception, and has dynamic environmental adaptability.

[0094] The method of the present invention will now be described in conjunction with the accompanying drawings and embodiments.

[0095] Figure 4 This is a schematic diagram of an optimal deployment method for multi-machine collaborative forest fire monitoring according to an embodiment of the present invention, as shown below. Figure 4 As shown, when the forest fire department receives a fire report, it estimates the location of the fire based on the report, sends the location to the UAV, and activates the UAV cluster for forest fire reconnaissance. The UAV obtains its own status information through its onboard navigation equipment (such as GPS and IMU) and environmental information through its onboard sensing equipment (such as cameras and anemometers). The UAV will move towards the estimated fire location. If it detects the fire boundary (i.e., the fire line), it will use the planning algorithm shown in formula (16) to achieve the optimal cluster deployment for forest fire monitoring. If the UAV does not detect the fire boundary, it will continue to move towards the location guided by the fire location until it detects the fire boundary.

[0096] This invention also provides an optimal deployment system for multi-machine collaborative forest fire monitoring, used to implement the above-mentioned optimal deployment method for multi-machine collaborative forest fire monitoring. The system includes:

[0097] The receiving module is used to receive fire reports;

[0098] The location estimation and model building module is used to estimate the fire location based on the fire report and build a forest fire spread grid model. The attributes of each grid cell in the forest fire spread grid model include combustible material type, terrain height, and combustion state. The combustion state includes unburned, incompletely burned, completely burned, and burnt out.

[0099] The startup module is used to send the fire location to the UAV and start the UAV cluster for forest fire reconnaissance;

[0100] The deployment module is used to set an objective function based on the importance of each block in the fire line and the quality of the fire scene images, and to determine the optimal coverage deployment of the UAV cluster for the fire scene based on the objective function; the fire line is composed of grid units in the incomplete combustion state, and each grid unit corresponds to a block in the fire line; the fire scene images are captured by cameras equipped with UAVs.

[0101] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the system described above can be found in the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0102] The present invention also provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors enable the aforementioned multi-machine collaborative optimal deployment method for forest fire monitoring.

[0103] The present invention also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enable the processor to implement the above-described multi-machine collaborative optimal deployment method for forest fire monitoring.

[0104] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0105] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0106] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0108] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

[0109] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. The above descriptions are exemplary and not exhaustive. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimal deployment of forest fire monitoring by multi-robot coordination, characterized in that, The method comprises: receiving a fire report; estimating a fire location and establishing a forest fire spread grid model according to the fire report; attributes of each grid cell in the forest fire spread grid model include fuel type, terrain height, and burning state; the burning state includes unburned, incomplete combustion, complete combustion, and burned out; sending the fire location to a UAV and starting a UAV cluster for forest fire reconnaissance; setting a target function according to the importance of each block in the fire line and the image quality of the fire field, and determining the optimal coverage deployment of the UAV cluster for the fire field according to the target function; the fire line is composed of grid cells with incomplete combustion state, and each grid cell corresponds to a block in the fire line; the fire field image is taken by a camera equipped on the UAV.

2. The method of claim 1, wherein, The importance of each block in the fireline is measured by a sensitivity function The measurement is made, The formula for the calculation is: ； wherein, represents a horizontal coordinate of a center point of a first block in the fire line, the first block being any block in the fire line; , , is a constant, and ; represents an included angle between an outer normal direction of the fire line at a point and a terrain gradient at a point ; represents an included angle between an outer normal direction of the fire line at a point and a wind direction; represents an importance degree of the first block, the greater the value of the importance degree of the first block, the higher the priority of the first block to be covered and investigated by the UAV.

3. The method of claim 2, wherein, The fire scene image quality is measured by a detection quality function is measured, The calculation formula is: ； in, Indicates the first in the UAV cluster The z-axis coordinates of each UAV; , These represent the minimum and maximum altitude limits for UAV flight, respectively. Indicates the first The set of all points within the field of view of the camera equipped by a UAV; Indicates the first The quality of the fire scene images obtained by one UAV capturing images of the first block. The higher the value, the higher the quality of the captured fire scene image.

4. The method of claim 3, wherein, The method for determining the optimal coverage deployment of the UAV cluster for the fire field according to the target function, comprises: According to the detection quality function Setting a cost function , The calculation formula is: ； wherein, a set of UAVs whose field of view covers the first bin; is a constant; denotes a cost when the first bin is covered by the field of view of multiple UAVs. According to the sensitivity function and the cost function Set the objective function , The calculation formula is: ； wherein represents live wire at the moment The optimal coverage deployment of the UAV cluster for the fire field is shown in the following formula: ； wherein, represents a position vector of the i-th UAV in the swarm of UAVs at the time instant t; represents a position vector of the i-th UAV in the swarm of UAVs at the time instant t; represents a position vector of the i-th UAV in the swarm of UAVs at the time instant t; represents a number of UAVs in the swarm of UAVs.

5. The method of claim 4, wherein, The method further comprises: Define the first in the UAV cluster The field of view of a UAV is The set of fire lines that are covered at all times is , The endpoint set is ,according to The objective function The integral is divided into three point sets, and the objective function after the division is... for: ； wherein denotes and the difference set of denotes and the difference set of According to the divided target function determining the horizontal control component and the height control component of the first UAV ； ； in, It is a constant; Indicates the first The horizontal position vectors of each UAV; Indicates in set Remove the first The set obtained from UAVs; Indicates the first Points on the boundary of the detection range of a UAV The unit outward normal vector; Represents a unit vector pointing vertically upwards; Used to control the first The movement of a UAV in the horizontal plane; Used to control the first The flight altitude of a UAV; According to , obtained : ； wherein, the coverage tracking behavior of the first UAV to the live wire is guided by the objective function .

6. The method of claim 5, wherein, The method further comprises: To avoid collision between UAVs, an inter-UAV collision avoidance potential field for a first UAV and a second UAV in a UAV cluster is defined as: ​​​ ； wherein, represents a position vector of the first UAV, represents a position vector of the first UAV, represents a position vector of the first UAV; represents a position vector of the first UAV; is a constant; is a preset distance; represents a modulus of a vector; Set representing a set of all UAVs in the UAV cluster whose distance to the :​​​​ ； wherein denotes the gradient operator; To maintain a safe distance of the UAV from the ground to prevent it from being burned by the flames, the safe distance potential field of the first UAV is defined as: :​ ； wherein, represents the z-axis coordinate of the first UAV; is a constant; represents a safety height of the UAV; To deriving the first UAV received ground repulsion force : 。 7. The method of claim 6, wherein, The method further comprises: According to , , obtaining a desired deployment location for the first UAV: ； ； wherein, represents a desired deployment position vector of the i-th UAV at time instant t; represents a first derivative of the position vector of the i-th UAV at time instant t; represents the mass of the UAV.​​​​ 8. A multi-machine cooperative forest fire monitoring optimal deployment system, characterized in that, The method further comprises: The method comprises: a receiving module configured to receive a fire report; a location estimation and model establishment module configured to estimate a fire location and establish a forest fire spread grid model according to the fire report; attributes of each grid cell in the forest fire spread grid model include fuel type, terrain height, and burning state; the burning state includes unburned, incomplete combustion, complete combustion, and burned out; a starting module configured to send the fire location to a UAV and start a UAV cluster for forest fire reconnaissance; 9. An electronic device, comprising: a deployment module configured to set a target function according to the importance of each block in the fire line and the image quality of the fire field, and determine the optimal coverage deployment of the UAV cluster for the fire field according to the target function; the fire line is composed of grid cells with incomplete combustion state, and each grid cell corresponds to a block in the fire line; the fire field image is taken by a camera equipped on the UAV. The method comprises: one or more processors; a memory configured to store one or more programs; 10. A computer-readable storage medium, characterized in that, wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-7. A computer readable storage medium having stored thereon executable instructions that, when executed by a processor, enable the processor to perform the method of any one of claims 1-7.

Images