Unmanned aerial vehicle group collaborative fire extinguishing intelligent management system

Abstract

The invention belongs to the technical field of intelligent management, and particularly discloses and provides an unmanned aerial vehicle group collaborative fire extinguishing intelligent management system, which comprises a flame position analysis module, a fire scene image is acquired by screening target unmanned aerial vehicles, and a flame area is identified to determine a fire source position. The fire behavior evaluation module divides a fire scene area according to the position of a fire source, monitors the wind speed, the combustible density and the flame temperature, analyzes a fire behavior spreading degree index, and screens out a fire behavior spreading area. The launching path analysis module screens fire extinguishing unmanned aerial vehicles based on unmanned aerial vehicle functions, and analyzes flight paths in combination with fire spreading areas, maps and unmanned aerial vehicle position information of a fire scene. The fire extinguishing agent throwing module collects data such as vertical temperature distribution, analyzes the target height and throwing distance of the fire extinguishing unmanned aerial vehicle, judges whether the target height and throwing distance are qualified or not and conducts adjustment. And the coverage area qualification judgment module is used for collecting a put fire extinguishing agent image, analyzing a coverage area qualification degree index, judging whether the fire extinguishing agent is qualified or not and performing early warning.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent management, and relates to an intelligent management system for coordinated firefighting by a swarm of unmanned aerial vehicles. Background Art

[0002] With the rapid development of science and technology and the acceleration of urbanization, fire prevention and control and emergency rescue work are facing increasingly greater challenges. Traditional fire-fighting methods are often limited by manpower, material resources and environmental complexity, making it difficult to respond to fire incidents quickly and accurately in the first place, especially in special scenarios such as high-rise buildings, forest fires, and petrochemicals. The difficulty of firefighting is significantly increased. Therefore, the development of an efficient, intelligent and collaborative fire-fighting management system has become an urgent problem to be solved.

[0003] For example, a cluster UAV forest fire extinguishing method with Chinese patent publication number CN113426044A includes the following steps: S101, constructing a single UAV fire extinguishing mounting device; including a mounting plate, a left fire extinguishing box, a middle fire extinguishing box, a right fire extinguishing box, a fire extinguishing box bottom plate, a fire extinguishing box rotating shaft, and a fire extinguishing box electromagnetic lock; S102, multi-machine collaborative forest fire extinguishing path planning and implementation; including constructing a UAV collaborative fire extinguishing path planning neural network, establishing a corresponding relationship between forest fire intelligence parameters and UAV cluster fire extinguishing path planning, obtaining planning and implementing fire extinguishing operations; S103, cluster UAV supplementary fire extinguishing implementation; including setting parameters, grid division, cluster flight planning, and implementing cluster UAV supplementary fire extinguishing planning; through the execution and implementation of the above steps, the cluster UAV forest fire extinguishing effect is achieved.

[0004] The above existing technologies have the following problems: 1. The existing technologies often only determine the target height of the UAV based on simple rules or experience, and lack scientific analysis methods. For example, the height of the UAV may only be adjusted according to a preset fixed height range without considering the complexity of the temperature distribution at the fire scene. This may lead to an unreasonable height for the UAV to drop the fire extinguisher, and it is impossible to achieve the ideal fire extinguishing effect while ensuring the safety of the UAV.

[0005] 2. Existing technologies have limited real-time monitoring capabilities for key parameters such as wind speed, combustible material distribution and temperature at the fire scene. The real-time changes in these parameters are crucial for determining the trend of fire spread and formulating fire-fighting strategies, but existing technologies often cannot provide sufficiently accurate and timely data support. Summary of the invention

[0006] In view of this, in order to solve the problems raised in the above background technology, an intelligent management system for coordinated fire fighting of drone swarms is proposed.

[0007] The objective of the present invention can be achieved through the following technical solutions: The present invention provides an intelligent management system for collaborative fire extinguishing of a drone swarm, including: a flame position analysis module: used to screen out target drones based on the functions of each drone in the drone swarm, collect images of the fire scene, identify the flame area in the images of the fire scene, and then analyze and obtain the fire source location in the fire scene.

[0008] Fire assessment module: used to divide the fire scene into various areas based on the location of the fire source in the fire scene, monitor the wind speed, combustible density and flame temperature of each area of ​​the fire scene in real time, analyze the fire spread index of each area of ​​the fire scene, determine whether each area of ​​the fire scene is a fire spread area, and then screen out the fire spread areas of the fire scene.

[0009] Deployment path analysis module: used to screen out fire-fighting drones based on the functions of each drone in the drone swarm, and then analyze the flight paths of each fire-fighting drone at the fire scene based on the fire spread areas at the fire scene, the fire scene map extracted from the database, and the location information of each fire-fighting drone collected.

[0010] Fire extinguishing agent delivery module: used to collect vertical temperature distribution data from the bottom of each fire-fighting drone to the ground, analyze the target height of each fire-fighting drone based on the maximum safe temperature that the fire-fighting drone can withstand extracted from the database, and then collect the wind speed and wind direction data of the environment in which each fire-fighting drone is located, as well as the flight speed of each fire-fighting drone, the initial injection velocity and injection angle of the fire-fighting agent, analyze the target horizontal position of each fire-fighting drone for releasing the fire-fighting agent, and then based on the collected horizontal position of each fire-fighting drone for releasing the fire-fighting agent, judge whether the horizontal position of each fire-fighting drone for releasing the fire-fighting agent is qualified, and then make adjustments.

[0011] Coverage area qualification judgment module: used to collect images of fire extinguishing agents dropped by each fire-fighting drone, analyze the coverage area qualification index of the fire-fighting agent dropped by each fire-fighting drone, determine whether the coverage area of ​​the fire-fighting agent dropped by each fire-fighting drone is qualified, and then issue an early warning.

[0012] Database: used to store the maximum safe temperature that fire-fighting drones can withstand and store fire scene maps.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By screening out target drones based on the functions of each drone in the drone swarm, the present invention can ensure that the most suitable drone for performing fire monitoring tasks is selected, collect images of the fire scene, identify the flame area in the images of the fire scene, and then analyze the location of the fire source in the fire scene, which can improve the collaborative operation and rapid response capabilities of the drone swarm, making fire-fighting operations more rapid and efficient, and helping to reduce the losses caused by fire.

[0014] (2) The present invention divides the fire scene into various areas based on the location of the fire source in the fire scene, monitors the wind speed, combustible density and flame temperature of each area of ​​the fire scene in real time, analyzes the fire spread index of each area of ​​the fire scene, determines whether each area of ​​the fire scene is a fire spread area, and then screens out each fire spread area of ​​the fire scene. The tasks of the drone group can be reasonably allocated, and different drones go to different fire spread areas according to the planned path, realizing multi-drone collaborative firefighting, avoiding the situation where multiple drones are concentrated in the same area while no one is extinguishing the fire in other areas, and giving full play to the overall firefighting efficiency of the drone group.

[0015] (3) The present invention collects vertical temperature distribution data from the bottom of each fire-fighting drone to the ground, and analyzes the target height of each fire-fighting drone based on the maximum safe temperature that the fire-fighting drone can withstand extracted from the database. This can avoid problems such as electronic component failure and body material damage caused by overheating of the fire-fighting drone, extend the service life of the fire-fighting drone, and reduce the risk of interruption of fire-fighting operations due to fire-fighting drone failure. In addition, by analyzing the target horizontal distance of each fire-fighting drone for releasing the fire-extinguishing agent, it is determined whether the target horizontal distance of each fire-fighting drone for releasing the fire-extinguishing agent is qualified, and then an early warning is issued. When the distance at which the fire-fighting drone releases the fire-extinguishing agent is too far or too close and does not meet the requirements, an early warning can be issued in time to allow the operator to respond quickly, thereby ensuring the efficient implementation of the fire-fighting operation and improving the success rate of fire-fighting.

[0016] (4) The present invention analyzes the coverage area qualification index of the fire extinguishing agent released by each fire-fighting drone based on the flight speed of each fire-fighting drone and the collected images of the fire-fighting drone releasing the fire-fighting agent, determines whether the coverage area of ​​the fire-fighting agent released by each fire-fighting drone is qualified, and then issues an early warning. It can accurately quantify the coverage effect of the fire-fighting agent. By analyzing the coverage area qualification index, it can be clearly known that the gap between the actual coverage of the fire-fighting agent released by each fire-fighting drone and the expected standard. The early warning mechanism can prompt the operator to respond quickly to unqualified situations to avoid the spread of fire or delay in fire extinguishing due to insufficient coverage of the fire-fighting agent. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention.

[0019] Figure 2The figure is a schematic diagram of the steps for analyzing the fire spread index of the fire scene area in the system of the present invention.

[0020] Figure 3 The present invention is a schematic diagram of the steps for determining whether the horizontal position of the fire extinguishing agent released by the fire-fighting drone is qualified in the system of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1 As shown, the present invention provides an intelligent management system for cooperative fire extinguishing of drone swarms, including: a flame position analysis module, a fire intensity assessment module, a delivery path analysis module, a fire extinguishing agent delivery module, a coverage area qualification judgment module and a database.

[0023] The fire assessment module is respectively connected to the flame position analysis module, the delivery path analysis module and the database; the delivery path analysis module is respectively connected to the fire assessment module, the fire extinguishing agent delivery module and the database; the fire extinguishing agent delivery module is respectively connected to the delivery path analysis module, the coverage area qualified judgment module and the database; the database is respectively connected to the fire assessment module, the delivery path analysis module and the fire extinguishing agent delivery module.

[0024] Flame position analysis module: used to screen out target drones based on the functions of each drone in the drone swarm, collect images of the fire scene, identify the flame area in the images of the fire scene, and then analyze the fire source location in the fire scene.

[0025] Furthermore, the analysis obtains the flame position in the fire scene, including: extracting the color characteristics, shape characteristics and dynamic changes of the flame from the image of the fire scene through image processing technology, and then combining with a deep learning algorithm to identify the flame area in the image of the fire scene, and then obtaining the position coordinates of each pixel point in the flame area in the image of the fire scene.

[0026] The camera on the target UAV is calibrated to obtain the internal and external parameters of the camera on the target UAV, and the real-time position coordinates of the target UAV are obtained through the GPS on the target UAV.

[0027] The internal and external parameters of the camera carried by the target UAV and the real-time position coordinates are used to convert the position coordinates of each pixel point in the flame area of ​​the fire scene image into the actual area coordinates of each pixel point in the flame area of ​​the fire scene through the principle of projection geometry.

[0028] The actual area coordinates of each pixel point in the flame area of ​​the fire scene are substituted into the average calculation formula to obtain the centroid coordinates of the flame area of ​​the fire scene, which are used as the fire source position of the fire scene.

[0029] As a preferred embodiment, the specific process of screening out target drones based on the functions of each drone in the drone swarm is as follows: based on the functions of each drone in the drone swarm, the drones can be divided into reconnaissance type, fire-fighting type, and auxiliary type, and any reconnaissance type drone will be screened out from each reconnaissance type drone as the target drone.

[0030] The specific process of extracting the color features, shape features and dynamic changes of flames from the images of the fire scene by image processing technology, and then combining with the deep learning algorithm to identify the flame area in the image of the fire scene is as follows: first, the fire scene images are collected by unmanned aerial vehicles and other equipment, and the images are manually annotated and divided into training sets, validation sets and test sets. Then, image enhancement and normalization processing are performed, and then features are extracted from three aspects: color, shape, and dynamic changes. The color features are obtained by converting the color space and calculating the statistical features of different channels; the shape features use the edge detection algorithm to extract edge information and calculate geometric features; the dynamic change features calculate the expansion and contraction speed and flickering frequency of the flame area for continuous multiple frames of images. Then, a suitable deep learning model such as a convolutional neural network is selected to construct a network structure, the annotated training set images are input for model training, a suitable optimization algorithm and learning rate are used, and the model hyperparameters and structure are adjusted in combination with the validation set evaluation. Finally, the image to be identified is input into the trained model for prediction, the output result is post-processed and verified and optimized in combination with other sensor data and prior knowledge, so as to identify the flame area in the fire scene image.

[0031] The present invention selects target drones based on the functions of each drone in the drone swarm, thereby ensuring that the most suitable drone for performing fire monitoring tasks is selected, collecting images of the fire scene, identifying the flame area in the images of the fire scene, and then analyzing the location of the fire source in the fire scene. This can improve the collaborative operation and rapid response capabilities of the drone swarm, making fire-fighting operations more rapid and efficient, and helping to reduce losses caused by fires.

[0032] Fire assessment module: used to divide the fire scene into various areas based on the location of the fire source in the fire scene, monitor the wind speed, combustible density and flame temperature of each area of ​​the fire scene in real time, analyze the fire spread index of each area of ​​the fire scene, determine whether each area of ​​the fire scene is a fire spread area, and then screen out the fire spread areas of the fire scene.

[0033] As a preferred embodiment, the specific process of dividing the fire scene into various areas based on the location of the fire source in the fire scene is: taking the flame position in the fire scene as the central reference point for dividing the area, taking the flame position in the fire scene as the center of the circle, and making concentric circles with different distances as the radius to obtain the areas of each concentric circle, which are used as the areas of the fire scene.

[0034] See also Figure 2 As shown, further, the analysis of the fire spread index of each area of ​​the fire scene includes: A1, arranging each monitoring point in each area of ​​the fire scene based on the principle of monitoring point layout, and then monitoring the wind speed and flame temperature of each monitoring point in each area of ​​the fire scene in real time through wind speed sensors and temperature sensors.

[0035] A2. Substitute the wind speed of each monitoring point in each area of ​​the fire scene into the average calculation formula to obtain the wind speed of each area of ​​the fire scene, recorded as wind g , where g represents the number corresponding to the g-th area of ​​the fire scene, g = 1, 2, 3, ...G.

[0036] A3. Based on the analysis of wind speed in each area of ​​the fire scene, the flame temperature of each area of ​​the fire scene is obtained in the same way, recorded as tem g .

[0037] A4. Use remote sensing technology to obtain the total mass and area of ​​combustibles in each area of ​​the fire scene. Divide the total mass of combustibles in each area of ​​the fire scene by the area to obtain the density of combustibles in each area of ​​the fire scene, recorded as density g .

[0038] As a preferred embodiment, the specific process of obtaining the total mass and area of ​​combustibles in each area of ​​the fire scene by remote sensing technology is as follows: after obtaining images of each area of ​​the fire scene by remote sensing technology, the spectral characteristics of different combustibles are then used to identify and classify the combustibles using a classification algorithm. For vegetation combustibles, a biomass estimation model is established, such as using the empirical relationship between the normalized vegetation index and biomass to estimate its mass; for non-vegetation combustibles, the mass is estimated based on its physical properties such as density and area, thereby obtaining the total mass of combustibles in each area of ​​the fire scene; based on the images of each area of ​​the fire scene, the images of each combustible in each area of ​​the fire scene are identified, and the area of ​​each combustible in each area of ​​the fire scene is obtained by image processing technology, thereby obtaining the area of ​​each combustible in each area of ​​the fire scene.

[0039] A5. Analyze the fire spread index of each area at the fire scene g .

[0040] The fire spread index of each area of ​​the fire scene g The specific calculation formula is: Wherein wind′, density′ and tem′ represent the allowable wind speed, allowable combustible density and allowable flame temperature of the set fire scene area respectively, β1, β2 and β3 represent the weights corresponding to the wind speed, combustible density and flame temperature of the fire scene area respectively, and β1+β2+β3=1.

[0041] As a preferred embodiment, the β1, β2 and β3 can be set to 0, 4, 0, 3 and 0, 3 respectively. Wind speed is one of the key factors for the spread of fire. A higher wind speed can accelerate the spread of flames and may aggravate the spread of fire. The density of combustibles determines the potential fuel supply at the fire scene. The higher the density of combustibles, the greater the possibility of fire spread. The flame temperature reflects the heat energy release at the fire scene. A higher flame temperature means more heat energy release, which may aggravate the spread of fire. Therefore, the weight corresponding to the wind speed in the fire scene area is greater than the weights corresponding to the combustible density and the flame temperature.

[0042] Furthermore, the determination of whether each area of ​​the fire scene is a fire spread area includes: comparing the fire spread degree index of each area of ​​the fire scene with a preset fire spread degree index threshold; if the fire spread degree index of an area of ​​the fire scene is greater than the preset fire spread degree index, then the area of ​​the fire scene is a fire spread area, thereby obtaining the fire spread areas of the fire scene.

[0043] The present invention divides the fire scene into various areas based on the location of the fire source in the fire scene, monitors the wind speed, combustible density and flame temperature of each area of ​​the fire scene in real time, analyzes the fire spread index of each area of ​​the fire scene, determines whether each area of ​​the fire scene is a fire spread area, and then screens out the fire spread areas of the fire scene. The present invention can reasonably allocate the tasks of the drone swarm, and different drones go to different fire spread areas according to the planned paths, realizing multi-drone collaborative firefighting, avoiding the situation where multiple drones are concentrated in the same area while no one is extinguishing fires in other areas, and giving full play to the overall firefighting efficiency of the drone swarm.

[0044] Deployment path analysis module: used to screen out fire-fighting drones based on the functions of each drone in the drone swarm, and then analyze the flight paths of each fire-fighting drone at the fire scene based on the fire spread areas at the fire scene, the fire scene map extracted from the database, and the location information of each fire-fighting drone collected.

[0045] Furthermore, the analysis of the flight paths of the fire-fighting drones at the fire scene includes: obtaining the position coordinates of the fire-fighting drones and the center position coordinates of the fire-spreading areas at the fire scene through GPS, substituting them into the Euclidean distance calculation formula, and obtaining the distance from the fire-fighting drones to the center position of the fire-spreading areas at the fire scene, which is recorded as L mh , where m represents the number corresponding to the mth fire-fighting drone, m=1, 2, 3, ...M, and h represents the number corresponding to the hth fire spread area, h=1, 2, 3, ...G.

[0046] Based on the fire spread index of each area at the fire scene, the fire spread index corresponding to each fire spread area at the fire scene is obtained, which is recorded as R h .

[0047] Based on the fire spread degree index corresponding to each fire spread area at the fire scene and the distance from each fire-fighting drone to the center position of each fire spread area at the fire scene, the target fire spread area of ​​each fire-fighting drone arriving at the fire scene is analyzed.

[0048] As a preferred embodiment, the specific process of analyzing the target fire spread area of ​​each fire-fighting drone arriving at the fire scene is: analyzing the priority index of each fire spread area of ​​each fire-fighting drone arriving at the fire scene mh , Where L′ represents the allowable distance from the fire-fighting drone to the center of the fire spread area at the fire scene.

[0049] The specific setting process of the allowable distance from the fire-fighting drone to the center position of the fire spread area at the fire scene is: substitute the distance from each fire-fighting drone to the center position of each fire spread area at the fire scene into the average calculation formula to obtain the average distance from each fire-fighting drone to the center position of the fire spread area at the fire scene, and then substitute it into the average calculation formula to obtain the average distance from the fire-fighting drone to the center position of the fire spread area at the fire scene, and use it as the allowable distance from the fire-fighting drone to the center position of the fire spread area at the fire scene.

[0050] The priority indexes of the fire spread areas when each fire-fighting drone arrives at the fire scene are sorted from large to small, and the fire spread area corresponding to the first-ranked priority index is allocated to the corresponding fire-fighting drone, thereby obtaining the target fire spread area when each fire-fighting drone arrives at the fire scene.

[0051] The straight line path of each fire-fighting drone arriving at the target fire spread area at the fire scene is used as the flight path of each fire-fighting drone at the fire scene.

[0052] Fire extinguishing agent delivery module: used to collect vertical temperature distribution data from the bottom of each fire-fighting drone to the ground, analyze the target height of each fire-fighting drone based on the maximum safe temperature that the fire-fighting drone can withstand extracted from the database, and then collect the wind speed and wind direction data of the environment in which each fire-fighting drone is located, as well as the flight speed of each fire-fighting drone, the initial injection velocity and injection angle of the fire-fighting agent, analyze the target horizontal position of each fire-fighting drone for releasing the fire-fighting agent, and then based on the collected horizontal position of each fire-fighting drone for releasing the fire-fighting agent, judge whether the horizontal position of each fire-fighting drone for releasing the fire-fighting agent is qualified, and then make adjustments.

[0053] See also Figure 3 As shown, further, the analysis of the target horizontal distance for each fire-fighting drone to release the fire extinguishing agent includes: collecting vertical temperature distribution data from the bottom of each fire-fighting drone to the ground through an infrared temperature sensor, and the vertical temperature distribution data from the bottom of each fire-fighting drone to the ground includes each height from the bottom of each fire-fighting drone to the ground and its corresponding temperature value.

[0054] The maximum safe temperature that the fire-fighting drone can withstand is compared with the temperature values ​​corresponding to each height from the bottom of each fire-fighting drone to the ground. If the temperature value corresponding to a certain height from the bottom of a fire-fighting drone to the ground is less than the maximum safe temperature that the fire-fighting drone can withstand, and the temperature value corresponding to the height below this height is greater than the maximum safe temperature that the fire-fighting drone can withstand, then the height from the bottom of the fire-fighting drone to the ground is set as the target height of the fire-fighting drone, and then the target height of each fire-fighting drone is obtained, which is recorded as H. m .

[0055] The wind speed and wind direction sensors carried by each fire-fighting drone collect the wind speed and wind direction angle of the environment in which each fire-fighting drone is located in real time, which are recorded as F m and θ m .

[0056] As a preferred embodiment, the wind direction angle of the environment in which each fire-fighting drone is located is specifically: the angle between the wind direction of the environment in which each fire-fighting drone is located and the flight direction of the corresponding fire-fighting drone.

[0057] The flight speed, initial velocity and injection angle of each fire-fighting UAV are obtained through the flight control platform of each fire-fighting UAV, which are recorded as v m , v′ m and α m ;

[0058] Analyze the target horizontal position x of each fire-fighting drone to release the fire extinguishing agent m , where t m It represents the time when the mth fire-fighting drone releases the fire extinguishing agent, and g represents the acceleration of gravity.

[0059] See also Figure 3 As shown, further, the specific process of judging whether the horizontal position of each fire-fighting drone dropping the fire extinguishing agent is qualified is as follows: B1. The horizontal position of each fire-fighting drone dropping the fire extinguishing agent is collected through GPS, and the target horizontal position of each fire-fighting drone dropping the fire extinguishing agent is compared with the horizontal position of the corresponding fire-fighting drone dropping the fire extinguishing agent, and the difference between the target horizontal position of each fire-fighting drone dropping the fire extinguishing agent and the horizontal position of the corresponding fire-fighting drone dropping the fire extinguishing agent is obtained, and recorded as the horizontal position distance difference of each fire-fighting drone dropping the fire extinguishing agent.

[0060] B2. Compare the horizontal position distance difference of each fire-fighting drone dropping the fire-extinguishing agent with the preset distance difference range. If the horizontal position distance difference of a fire-fighting drone dropping the fire-extinguishing agent does not fall within the preset distance difference range, the horizontal position of the fire-fighting drone dropping the fire-extinguishing agent is unqualified and is adjusted.

[0061] As a preferred embodiment, the specific process of adjusting is: moving the horizontal position of each fire-fighting drone dropping the fire-extinguishing agent in the direction of the target horizontal position of the corresponding fire-fighting drone dropping the fire-extinguishing agent.

[0062] The present invention collects vertical temperature distribution data from the bottom of each fire-fighting drone to the ground, analyzes the target height of each fire-fighting drone based on the maximum safe temperature that the fire-fighting drone can withstand extracted from a database, and can avoid problems such as electronic component failure and body material damage caused by overheating of the fire-fighting drone, thereby extending the service life of the fire-fighting drone and reducing the risk of interruption of fire-fighting operations due to fire-fighting drone failure. In addition, by analyzing the target horizontal distance for each fire-fighting drone to release the fire-extinguishing agent, it is determined whether the target horizontal distance for each fire-fighting drone to release the fire-extinguishing agent is qualified, and then an early warning is issued. When the distance for the fire-fighting drone to release the fire-extinguishing agent is too far or too close and does not meet the requirements, an early warning can be issued in time to allow the operator to respond quickly, thereby ensuring the efficient implementation of the fire-fighting operation and improving the success rate of fire-fighting.

[0063] Coverage area qualification judgment module: used to collect images of fire extinguishing agents dropped by each fire-fighting drone, analyze the coverage area qualification index of the fire-fighting agent dropped by each fire-fighting drone, determine whether the coverage area of ​​the fire-fighting agent dropped by each fire-fighting drone is qualified, and then issue an early warning.

[0064] Furthermore, the analysis of the coverage area qualification index of the fire extinguishing agent includes: collecting images of the fire extinguishing agent released by each fire extinguishing drone through a high-speed camera carried by each fire extinguishing drone, and using image recognition technology and area calculation algorithm to determine the coverage area of ​​the fire extinguishing agent released by each fire extinguishing drone, which is recorded as U m .

[0065] The analysis of the coverage area of ​​each fire-fighting drone dropping fire extinguishing agent is qualified m , Where U′ represents the standard coverage area of ​​the fire-fighting drone for releasing fire-extinguishing agents.

[0066] Furthermore, the specific process of judging whether the coverage area of ​​the fire extinguishing agent is qualified is as follows: comparing the coverage area qualification index of the fire extinguishing agent dropped by each fire-fighting drone with the preset coverage area qualification index threshold; if the coverage area qualification index of the fire extinguishing agent dropped by a certain fire-fighting drone is less than the preset coverage area qualification index threshold, then the coverage area of ​​the fire-fighting agent dropped by the fire-fighting drone is unqualified, and then each fire-fighting drone that has unqualified fire-fighting agent drop is given an early warning.

[0067] The present invention analyzes the coverage area qualification index of the fire extinguishing agent dropped by each fire extinguishing drone based on the flight speed of each fire extinguishing drone and the collected images of the fire extinguishing agent dropped by each fire extinguishing drone, determines whether the coverage area of ​​the fire extinguishing agent dropped by each fire extinguishing drone is qualified, and then issues an early warning. The present invention can accurately quantify the coverage effect of the fire extinguishing agent. By analyzing the coverage area qualification index, the gap between the actual coverage of the fire extinguishing agent dropped by each fire extinguishing drone and the expected standard can be clearly known. The early warning mechanism can prompt the operator to respond quickly to unqualified situations, thereby avoiding the spread of fire or delay in fire extinguishing due to insufficient coverage of the fire extinguishing agent.

[0068] Database: used to store the maximum safe temperature that fire-fighting drones can withstand and to store and borrow fire scene maps.

[0069] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. An intelligent management system for cooperative firefighting by drone swarms, characterized by: include: Flame location analysis module: used to select target drones based on the functions of each drone in the drone group, collect images of the fire scene, identify the flame area in the images of the fire scene, and then analyze the fire source location in the fire scene; Fire assessment module: used to divide the fire scene into various areas based on the location of the fire source in the fire scene, monitor the wind speed, combustible density and flame temperature of each area of ​​the fire scene in real time, analyze the fire spread index of each area of ​​the fire scene, determine whether each area of ​​the fire scene is a fire spread area, and then screen out each fire spread area of ​​the fire scene; The delivery path analysis module is used to select the fire-fighting drones based on the functions of each drone in the drone group, and then analyze the flight paths of each fire-fighting drone at the fire scene based on the fire spread areas at the fire scene, the fire scene map extracted from the database, and the location coordinates of each fire-fighting drone collected; Fire extinguishing agent delivery module: used to collect vertical temperature distribution data from the bottom of each fire-fighting drone to the ground, analyze the target height of each fire-fighting drone based on the maximum safe temperature that the fire-fighting drone can withstand extracted from the database, and then collect the wind speed and wind direction data of the environment in which each fire-fighting drone is located, as well as the flight speed of each fire-fighting drone, the initial injection velocity and injection angle of the fire-fighting agent, analyze the target horizontal position of each fire-fighting drone to release the fire-fighting agent, and then judge whether the horizontal position of each fire-fighting drone to release the fire-fighting agent is qualified based on the collected horizontal position of each fire-fighting drone to release the fire-fighting agent, and then make adjustments; Coverage area qualification judgment module: used to collect images of fire extinguishing agents dropped by each fire-fighting drone, analyze the coverage area qualification index of the fire-fighting agents dropped by each fire-fighting drone, determine whether the coverage area of ​​the fire-fighting agents dropped by each fire-fighting drone is qualified, and then issue an early warning; Database: used to store the maximum safe temperature that fire-fighting drones can withstand and store fire scene maps.

2. The intelligent management system for cooperative firefighting of drone swarms according to claim 1 is characterized by: The analysis obtains the flame position in the fire scene, including: The color features, shape features and dynamic changes of flames are extracted from the images of the fire scene through image processing technology, and then combined with deep learning algorithms, the flame area in the images of the fire scene is identified, and the position coordinates of each pixel point in the flame area in the images of the fire scene are obtained; Calibrate the camera on the target UAV to obtain the internal and external parameters of the camera on the target UAV, and obtain the real-time position coordinates of the target UAV through the GPS on the target UAV; The internal parameters and external parameters of the camera carried by the target UAV and the real-time position coordinates are used to convert the position coordinates of each pixel point in the flame area of ​​the fire scene image into the actual area coordinates of each pixel point in the flame area of ​​the fire scene through the principle of projection geometry; The actual area coordinates of each pixel point in the flame area of ​​the fire scene are substituted into the average calculation formula to obtain the centroid coordinates of the flame area of ​​the fire scene, which are used as the fire source position of the fire scene.

3. The intelligent management system for cooperative firefighting of drone swarms according to claim 1 is characterized by: The fire spread index of each area of ​​the fire scene is analyzed, including: A1. Based on the principle of monitoring point layout, various monitoring points are arranged in various areas of the fire scene, and then the wind speed and flame temperature of each monitoring point in each area of ​​the fire scene are monitored in real time through wind speed sensors and temperature sensors; A2. Substitute the wind speed of each monitoring point in each area of ​​the fire scene into the average calculation formula to obtain the wind speed of each area of ​​the fire scene, recorded as wind g , where g represents the number corresponding to the g-th area of ​​the fire scene, g = 1, 2, 3, ... G; A3. Based on the analysis of wind speed in each area of ​​the fire scene, the flame temperature of each area of ​​the fire scene is obtained in the same way, recorded as tem g ; A4. Use remote sensing technology to obtain the total mass and area of ​​combustibles in each area of ​​the fire scene. Divide the total mass of combustibles in each area of ​​the fire scene by the area to obtain the density of combustibles in each area of ​​the fire scene, recorded as density g . A5. Analyze the fire spread index of each area at the fire scene g .

4. The intelligent management system for cooperative firefighting of drone swarms according to claim 1 is characterized by: The step of judging whether each area of ​​the fire scene is a fire spreading area includes: The fire spread index of each area of ​​the fire scene is compared with the preset fire spread index threshold. If the fire spread index of an area of ​​the fire scene is greater than the preset fire spread index, then the area of ​​the fire scene is a fire spread area, and then the fire spread areas of the fire scene are obtained.

5. The intelligent management system for cooperative firefighting of drone swarms according to claim 1 is characterized by: The analysis of the flight paths of each drone at the fire scene includes: The position coordinates of each fire-fighting drone and the center coordinates of each fire spread area at the fire scene are obtained through GPS, and they are substituted into the Euclidean distance calculation formula to obtain the distance from each fire-fighting drone to the center of each fire spread area at the fire scene, which is recorded as L mh , where m represents the number corresponding to the m-th fire-fighting drone, m=1, 2, 3, ...M, and h represents the number corresponding to the h-th fire spreading area, h=1, 2, 3, ...G; Based on the fire spread index of each area at the fire scene, the fire spread index corresponding to each fire spread area at the fire scene is obtained, which is recorded as R h ; Based on the fire spread index corresponding to each fire spread area at the fire scene and the distance from each fire-fighting drone to the center of each fire spread area at the fire scene, the target fire spread area of ​​each fire-fighting drone arriving at the fire scene is analyzed; The straight line path of each fire-fighting drone arriving at the target fire spread area at the fire scene is used as the flight path of each fire-fighting drone at the fire scene.

6. The intelligent management system for cooperative firefighting of drone swarms according to claim 5 is characterized by: The analysis of the target height of each firefighting drone includes: The vertical temperature distribution data from the bottom of each fire-fighting drone to the ground is collected by an infrared temperature sensor. The vertical temperature distribution data from the bottom of each fire-fighting drone to the ground includes each height from the bottom of each fire-fighting drone to the ground and its corresponding temperature value; The maximum safe temperature that the fire-fighting drone can withstand is compared with the temperature values ​​corresponding to each height from the bottom of each fire-fighting drone to the ground. If the temperature value corresponding to a certain height from the bottom of a fire-fighting drone to the ground is less than the maximum safe temperature that the fire-fighting drone can withstand, and the temperature value corresponding to the height below this height is greater than the maximum safe temperature that the fire-fighting drone can withstand, then the height from the bottom of the fire-fighting drone to the ground is set as the target height of the fire-fighting drone, and then the target height of each fire-fighting drone is obtained, which is recorded as H. m .

7. The intelligent management system for coordinated firefighting of drone swarms according to claim 6 is characterized by: The analysis of the target horizontal position of each fire-fighting drone dropping the fire-extinguishing agent includes: The wind speed and wind direction sensors carried by each fire-fighting drone collect the wind speed and wind direction angle of the environment in which each fire-fighting drone is located in real time, which are denoted as F m and θ m ; The flight speed, initial velocity and injection angle of each fire-fighting UAV are obtained through the flight control platform of each fire-fighting UAV, which are recorded as v m , v′ m and α m ; Analyze the target horizontal position x of each fire-fighting drone to release the fire extinguishing agent m , where t m It represents the time when the mth fire-fighting drone releases the fire extinguishing agent, and g represents the acceleration of gravity.

8. The intelligent management system for cooperative firefighting of drone swarms according to claim 1 is characterized by: The specific process of judging whether the horizontal position of each fire-fighting drone dropping the fire-extinguishing agent is qualified is as follows: B1. Collect the horizontal position of each fire-fighting drone dropping the fire extinguishing agent through GPS, compare the target horizontal position of each fire-fighting drone dropping the fire extinguishing agent with the horizontal position of the corresponding fire-fighting drone dropping the fire extinguishing agent, and obtain the difference between the target horizontal position of each fire-fighting drone dropping the fire extinguishing agent and the horizontal position of the corresponding fire-fighting drone dropping the fire extinguishing agent, which is recorded as the horizontal position distance difference of each fire-fighting drone dropping the fire extinguishing agent; B2. Compare the horizontal position distance difference of each fire-fighting drone dropping the fire-extinguishing agent with the preset distance difference range. If the horizontal position distance difference of a fire-fighting drone dropping the fire-extinguishing agent does not fall within the preset distance difference range, the horizontal position of the fire-fighting drone dropping the fire-extinguishing agent is unqualified and is adjusted.

9. The intelligent management system for cooperative firefighting of drone swarms according to claim 5 is characterized by: The analysis of the coverage area qualification index of each fire-fighting drone dropping fire-fighting agents includes: The high-speed camera carried by each fire-fighting drone collects images of the fire-fighting agent dropped by each fire-fighting drone, and uses image recognition technology and area calculation algorithm to determine the coverage area of ​​the fire-fighting agent dropped by each fire-fighting drone, which is recorded as U m ; The analysis of the coverage area of ​​each fire-fighting drone dropping fire extinguishing agent is qualified m , Where U′ represents the standard coverage area of ​​the fire-fighting drone for releasing fire-extinguishing agents.

10. The intelligent management system for cooperative firefighting of drone swarms according to claim 1 is characterized by: The specific process of judging whether the coverage area of ​​the fire extinguishing agent is qualified is as follows: The coverage area qualification index of the fire extinguishing agent released by each fire-fighting drone is compared with the preset coverage area qualification index threshold. If the coverage area qualification index of the fire-fighting drone is less than the preset coverage area qualification index threshold, the coverage area of ​​the fire-fighting drone that releases the fire-fighting agent is unqualified, and then the fire-fighting drones that fail to release the fire-fighting agent are warned.

Images