Unmanned aerial vehicle emergency management method and system, electronic equipment and storage medium

Abstract

The invention discloses an unmanned aerial vehicle emergency management method and system, electronic equipment and a storage medium. The method comprises the following steps: determining a disaster area and an emergency scene; matching a preset emergency strategy according to the emergency scene; dividing the range of the affected area and generating a plurality of affected area monitoring tasks; constructing a first unmanned aerial vehicle group, planning a first flight route of each unmanned aerial vehicle in the first unmanned aerial vehicle group by using a multi-target optimization algorithm, calling the unmanned aerial vehicles to go to the disaster-affected area to execute a monitoring task, obtaining real-time data of the disaster-affected area, and recognizing key point distribution of the disaster-affected area according to a preset emergency strategy, predicting the spreading direction and range of the affected area; determining rescue areas, dividing a plurality of rescue areas and generating a plurality of rescue tasks; and constructing a second unmanned aerial vehicle group, planning a second flight route of each unmanned aerial vehicle in the second unmanned aerial vehicle group by using a multi-objective optimization algorithm, and calling the unmanned aerial vehicles to go to the rescue area to execute the rescue task. According to the invention, the rescue efficiency and precision can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of drone management, and specifically to a drone emergency management method, system, electronic device and storage medium. Background Art

[0002] With the development of drone technology, drones are increasingly being used in emergency management, disaster relief and other fields. However, existing drone applications are mostly focused on a single function, such as simple material delivery or data collection, lacking a systematic management mechanism and intelligent dispatching capabilities, and are unable to fully respond to complex and changing emergency rescue needs.

[0003] In the existing technology, most drones used in emergency scenarios still rely on manual control or autonomous flight along preset paths, lacking dynamic task allocation and remote collaboration capabilities. This limitation leads to inefficiency in drone swarms when performing multi-task and cross-regional rescue, delayed information transmission, and insufficient response when facing emergencies, which seriously affects the effectiveness and safety of rescue.

[0004] Therefore, it is necessary to design a drone emergency management method to dynamically generate drone task allocation and collaborative execution in the face of a variety of complex and changeable emergency scenarios to improve the effectiveness and accuracy of rescue. Summary of the invention

[0005] In order to dynamically generate drone task allocation and collaborative execution in the face of various emergency scenarios and improve rescue efficiency and accuracy, the present application provides a drone emergency management method, system, electronic device and storage medium.

[0006] In a first aspect, the present application provides a drone emergency management method, comprising: Receive alarm information and preliminarily determine the disaster-affected area and emergency scenarios; the emergency scenarios include: fire, earthquake and chemical leakage; obtain data information of drones around the disaster-affected area; the data information includes location data and status data; According to the emergency scenario matching preset emergency strategies, each emergency scenario is provided with a preset emergency strategy matching it; the preset emergency strategy includes: predetermined rules for executing emergency tasks, and types of drones for executing emergency tasks; wherein the predetermined rules for executing emergency tasks include key points of the executed emergency tasks; Divide the disaster-stricken area and generate a number of disaster-stricken area monitoring tasks accordingly; construct a first drone group based on the data information of drones around the disaster-stricken area and the matched drone types, and use a multi-objective optimization algorithm to plan the first flight route of each drone in the first drone group, so that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized; call the corresponding drone to the disaster-stricken area according to the planned first flight route of each drone in the first drone group to perform the monitoring task and obtain real-time data of the disaster-stricken area; use image analysis technology to analyze the real-time data of the disaster-stricken area, identify the distribution of key points in the disaster-stricken area according to the predetermined rules for executing emergency tasks, and predict the spread direction and range of the disaster-stricken area; The rescue area is determined based on the distribution of key points in the identified disaster-stricken area and the predicted spreading direction and range of the disaster-stricken area, the rescue area is divided and a number of rescue tasks are generated accordingly; the drone data information around the rescue area is obtained, and based on the data information of the drones around the rescue area and the matched drone types, a second drone swarm is constructed and a second flight route for each drone in the second drone swarm is planned using a multi-objective optimization algorithm, so that each drone can complete the rescue task of a single rescue area and the total time to complete the rescue task is minimized; and the corresponding drone is called to the rescue area to perform the rescue task according to the planned second flight route of each drone in the second drone swarm.

[0007] By adopting the above scheme, taking into account that different emergency scenarios have different emphases on disaster monitoring and rescue, the corresponding type of drone is selected to complete the monitoring and rescue of the disaster-stricken area based on the specific emergency scenario, thereby improving the effectiveness of the rescue; based on the monitoring and analysis of the real-time data of the disaster-stricken area by the drone swarm, the changes in the disaster-stricken area in the future period are predicted, and rescue missions are dynamically generated. The routes of the drones in the drone swarm are planned according to the rescue missions to complete the coordinated rescue of the drones, thereby improving the effectiveness and accuracy of the rescue.

[0008] Preferably, it also includes: Obtaining the disaster level corresponding to each divided disaster area according to the received alarm information; further dividing the divided disaster area according to the obtained disaster level corresponding to each divided disaster area; wherein different disaster levels are preset with corresponding disaster area range division standards; The key point distribution of the identified disaster-stricken areas corresponding to each divided rescue area and the predicted spreading direction and range of the disaster-stricken areas are input into the disaster level judgment model corresponding to the emergency scenario to obtain the disaster level corresponding to each rescue area; there are multiple disaster level judgment models, each disaster level judgment model matches an emergency scenario and adopts a neural network model, which is generated by training the key point distribution of the identified historical disaster-stricken areas and the predicted spreading direction and range of the historical disaster-stricken areas corresponding to the historical rescue areas marked with disaster levels under the matched emergency scenarios; each divided rescue area is further divided according to the disaster level division output by each divided rescue area; wherein different disaster levels are preset with corresponding rescue area range division standards.

[0009] By adopting the above scheme, taking into account the different requirements for monitoring and rescue of different disaster levels in different emergency scenarios, the higher the disaster level, the more refined monitoring and rescue is needed. The disaster level of the disaster-stricken area and the rescue task is determined, and the division standards of the corresponding emergency scenarios and the corresponding disaster levels are adaptively matched. The monitoring and rescue areas are further divided to refine the monitoring and rescue and improve the accuracy of the rescue.

[0010] Preferably, it also includes: The acquired data information of drones around the disaster-stricken area also includes drone performance data, and the performance data includes power consumption data per kilometer; Considering the performance data of the drones, based on the data information of the drones around the disaster-stricken area and the matched drone types, the third drone group is reconstructed and the third flight route of each drone in the third drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the monitoring task of a single disaster-stricken area and the weighted value of the total time to complete the monitoring task and the total power consumption is minimized; the corresponding drone is called to go to the disaster-stricken area to perform the monitoring task according to the planned third flight route of each drone in the third drone group, instead of calling the corresponding drone to go to the disaster-stricken area to perform the monitoring task according to the planned first flight route of each drone in the first drone group; Taking into account the drone performance data, based on the data information of the drones around the rescue area and the matching drone types, the fourth drone swarm is reconstructed and the fourth flight route of each drone in the fourth drone swarm is planned using a multi-objective optimization algorithm, so that each drone can complete the rescue mission in a single rescue area and the weighted values ​​of the total time to complete the rescue mission and the total power consumption are minimized; according to the planned fourth flight route of each drone in the fourth drone swarm, the corresponding drone is called to go to the rescue area to perform the rescue mission instead of calling the corresponding drone to go to the disaster-stricken area to perform the rescue mission according to the planned second flight route of the second drone swarm.

[0011] By adopting the above solution, in the process of dynamically generating rescue tasks and dynamically adjusting and optimizing the paths according to the rescue tasks, comprehensive consideration is given from multiple angles such as minimizing the task completion time and resource consumption, thereby improving the effectiveness and accuracy of the rescue while maximizing resource utilization.

[0012] Preferably, it also includes: The flight status data of each drone in the first drone group or the second drone group is received in real time, and the flight status data of the drone is predicted to be faulty by using machine learning technology; if the prediction result is that there is a fault, a return instruction is generated for the drone with the fault; at the same time, the type and area of ​​the emergency task not executed are determined according to the record of the emergency task executed by the drone with the fault, and are recorded as the type of the area to be supplemented and the area to be supplemented; According to the type of emergency mission to be supplemented, a redundant drone of the corresponding type is selected from the original take-off position station of the faulty drone, and a supplementary emergency mission route is generated from the current location of the redundant drone to the area where the emergency mission is to be supplemented, and the supplementary emergency mission route does not conflict with the first flight route or the second flight route of the original faulty drone.

[0013] By adopting the above solution, considering using drone swarms to collaboratively complete the monitoring of disaster-stricken areas or the rescue of rescue areas, if a drone fails, a redundant drone is selected to continue to replace the failed drone to complete the mission, thereby improving the effectiveness and safety of rescue in the face of emergencies.

[0014] Preferably, it also includes: Receive the flight status data of each drone in the first drone group or the second drone group in real time, and determine whether the power of each drone reaches the minimum power preset threshold; if reached, send a hovering command to the drone whose power reaches the minimum power preset threshold, and call the energy replenishment drone, so that the energy replenishment drone arrives above the real-time position of the drone whose power reaches the minimum power preset threshold along the flight route of the drone whose power reaches the minimum power preset threshold, and after the energy replenishment drone completes the replenishment of the drone whose power reaches the minimum power preset threshold, return to the flight starting point along the flight route of the drone whose power reaches the minimum power preset threshold.

[0015] By adopting the above solution, considering using a swarm of drones to collaboratively complete the monitoring of the disaster-stricken area or the rescue process in the rescue area, if there are emergencies or other factors that cause the drones to run out of power, energy-supplying drones can be added to provide aerial charging services for the drones performing the mission, thereby improving rescue efficiency and safety.

[0016] Preferably, it also includes: When the UAV performs monitoring tasks in the disaster-stricken area or performs rescue tasks in the rescue area, the flight mode corresponding to the monitoring tasks in the disaster-stricken area or the rescue tasks in the rescue area is matched in advance according to the determined emergency scenario, and the monitoring tasks in the disaster-stricken area or the rescue tasks in the rescue area are completed according to the matched flight mode; wherein each emergency scenario is matched with a preset flight mode for the UAV to perform monitoring tasks and a flight mode for the UAV to perform rescue tasks.

[0017] By adopting the above solution, taking into account the different emphases of monitoring the disaster-stricken areas and rescue in the rescue areas in different emergency scenarios, appropriate flight modes are planned to improve the accuracy and efficiency of monitoring and rescue.

[0018] Preferably, the type of drones that perform emergency tasks in the preset emergency strategy according to the emergency scenario matching includes: The types of drones that perform emergency tasks in the preset emergency strategies that match fire emergency scenarios include thermal imaging drones, high-definition camera drones, and drones equipped with fire-fighting materials or fire-fighting devices; the types of drones that perform emergency tasks in the preset emergency strategies that match earthquake emergency scenarios include infrared detection drones, high-definition camera drones, and drones equipped with rescue supplies; the types of drones that perform emergency tasks in the preset emergency strategies that match chemical leak emergency scenarios include drones equipped with chemical sensors, high-definition camera drones, and drones equipped with chemical neutralization materials or chemical adsorption materials.

[0019] By adopting the above solution, taking into account the different emphases of different emergency scenarios on monitoring of disaster-stricken areas and rescue in rescue areas, drones equipped with different types of equipment are adaptively selected to meet the requirements of monitoring and rescue in different emergency scenarios.

[0020] In a second aspect, the present application provides a drone emergency management system, comprising: A disaster area and scene determination module is used to receive alarm information and preliminarily determine the disaster area and emergency scene; the emergency scene includes: fire, earthquake and chemical leakage; obtain data information of drones around the disaster area; the data information includes location data and status data; The disaster emergency strategy matching module is used to match the preset emergency strategy according to the emergency scenario. Each emergency scenario is provided with a preset emergency strategy matching it; the preset emergency strategy includes: predetermined rules for executing emergency tasks and types of drones for executing emergency tasks; wherein the predetermined rules for executing emergency tasks include key points of the executed emergency tasks; A disaster-stricken area monitoring module is executed to divide the disaster-stricken area and generate a number of disaster-stricken area monitoring tasks accordingly; based on the data information of the drones around the disaster-stricken area and the matched drone types, a first drone group is constructed and a first flight route of each drone in the first drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized; according to the planned first flight route of each drone in the first drone group, the corresponding drone is called to go to the disaster-stricken area to perform the monitoring task and obtain real-time data of the disaster-stricken area; the real-time data of the disaster-stricken area is analyzed using image analysis technology, and the distribution of key points in the disaster-stricken area is identified according to the predetermined rules for executing emergency tasks, and the spreading direction and range of the disaster-stricken area are predicted; The rescue module of the rescue area is executed, which is used to determine the rescue area according to the distribution of key points in the identified disaster-stricken area and the predicted spreading direction and range of the disaster-stricken area, divide the rescue area and generate a number of rescue tasks accordingly; obtain the drone data information around the rescue area, and build a second drone group based on the data information of the drones around the rescue area and the matching drone types, and use a multi-objective optimization algorithm to plan the second flight route of each drone in the second drone group, so that each drone can complete the rescue task of a single rescue area and the total time to complete the rescue task is minimized, and call the corresponding drone to the rescue area according to the planned second flight route of each drone in the second drone group to perform the rescue task.

[0021] By adopting the above scheme, the disaster emergency strategy matching module and the disaster area monitoring module are used to improve the accuracy of real-time data monitoring in the disaster area, and the rescue area rescue module is used to dynamically generate rescue tasks, so as to achieve dynamic coordination of drones in complex and changing emergency scenarios and improve the accuracy and effectiveness of rescue.

[0022] In a third aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method as described above.

[0023] In a fourth aspect, the present application provides a computer device, comprising a memory, a processor, and a program stored and executable on the memory, wherein the program implements the steps of the above method when executed by the processor.

[0024] In summary, this application has the following beneficial effects: 1. Considering that disaster monitoring and rescue have different focuses in different emergency scenarios, select corresponding types of drones to complete the monitoring and rescue of the disaster-stricken areas, improve the accuracy of monitoring data and the effectiveness of rescue; build a drone swarm based on the scope of the disaster-stricken area and use the drone swarm to monitor the disaster-stricken area and analyze the real-time data, predict the changes in the disaster-stricken area in the future period, dynamically generate rescue tasks, plan the routes of drones in the drone swarm according to the time or resource utilization requirements of the rescue task and the specific task execution process to complete the coordinated rescue of drones, and improve the effectiveness and accuracy of rescue; 2. Considering the different requirements for the degree of refinement of disaster monitoring and rescue in different emergency scenarios and disaster levels, determine the specific emergency scenarios and disaster levels, match the division standards of the preset disaster areas or rescue areas, generate corresponding monitoring or rescue tasks, and call drones to further perform monitoring and rescue; 3. Consider using drones to complete monitoring and rescue. In case of drone failure, design redundant drones. In case of insufficient power, design energy-replenishing drones to improve the effectiveness of rescue in the face of emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flowchart of the UAV emergency management method described in a specific embodiment; Figure 2 It is a structural schematic diagram of the UAV emergency management system described in a specific embodiment. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0027] like Figure 1 As shown, the embodiment of the present application discloses a first aspect, and the present application provides a method for emergency management of a drone, and the specific steps include: S1. Receive alarm information and preliminarily determine the disaster-affected areas and emergency scenarios.

[0028] Specifically, the alarm information about emergency situations sent by the alarm system is received, and the alarm information includes: emergency scenario information, location, scope, environment and disaster level information of the disaster-stricken area, etc.; wherein the emergency scenario information mainly includes typical emergency scenarios, such as: fire, earthquake and chemical leakage, etc.

[0029] S2. Obtain data information of drones around the disaster area.

[0030] According to the location of the disaster-stricken area, the information of drones around the disaster-stricken area is queried in combination with the recorded drone sites or the received drone communication information. The size of the disaster-stricken area can be determined according to the artificially set distance. The data information of drones around the disaster-stricken area finally obtained includes: drone model and number, drone location data, drone operation data, drone status data (idle or mission execution status) and drone performance data (such as computing resource data, power consumption per kilometer), etc.

[0031] S3. Match the preset emergency strategy according to the emergency scenario.

[0032] Specifically, each emergency scenario is provided with a preset emergency strategy matching it, such as: a first preset emergency strategy matching a fire emergency scenario, a second preset emergency strategy matching an earthquake emergency scenario, and a third preset emergency strategy matching a chemical leak emergency scenario.

[0033] Each matching preset emergency strategy has: predetermined rules for executing emergency tasks, and types of drones for executing emergency tasks; wherein, the predetermined rules for executing emergency tasks include typical types of emergency tasks for corresponding scenarios and key points of emergency tasks to be executed. For example, the key points of emergency tasks executed in the first preset emergency strategy include: key points such as fire source points, high temperature points, and boundary points, etc., and the types of drones for executing emergency tasks in the first preset emergency strategy include: thermal imaging drones, high-definition camera drones, and drones equipped with fire-fighting materials or fire-fighting devices; the key points of emergency tasks executed in the second preset emergency strategy include: disaster-affected boundary points, locations of trapped persons, etc., and the types of drones for executing emergency tasks in the second preset emergency strategy include: infrared detection drones, high-definition camera drones, and drones equipped with rescue materials; the key points of emergency tasks executed in the third preset emergency strategy include: chemical leakage sources, chemical high concentration points, and boundary points, etc., and the types of drones for executing emergency tasks in the third preset emergency strategy include: drones equipped with chemical sensors, high-definition camera drones, and drones equipped with chemical neutralization materials or chemical adsorption materials.

[0034] S4. Divide the disaster-stricken area and generate monitoring tasks for the disaster-stricken area. Plan the paths of the drones in the drone swarm according to the monitoring tasks for the disaster-stricken area and call the drone swarm to collaboratively perform the monitoring tasks.

[0035] Specifically, the scope of the disaster-stricken area can be preliminarily obtained based on the alarm information. In order to ensure the accuracy and timeliness of rescue in the disaster-stricken area, it is necessary to monitor the disaster-stricken area and then divide the disaster-stricken area in advance. The specific division can be based on the artificially preset area division of the single UAV cruise monitoring area to generate the corresponding disaster-stricken area monitoring task. Each area for single UAV cruise monitoring is regarded as a disaster-stricken area monitoring area.

[0036] The number of drones required is determined according to the number of monitoring tasks for the disaster-stricken area generated, and the type of drone required is determined according to the type of drone matched by the determined emergency scenario; in combination with the data information of the drones around the disaster-stricken area, drones that meet the drone type and are in an idle state and have sufficient power are pre-queried, a first drone set is generated, and drones that meet the required number are selected from the first drone set to construct a first drone group (the default number of surrounding drones is sufficient or the manually set distance can be adjusted to re-expand the surrounding range to obtain a sufficient number of drones); a first flight path of the first drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized; wherein, the process of determining the first drone group and planning the first path of each drone of the first drone group under the condition of meeting the restriction requirements includes: pre-traversing and forming multiple drone groups, and continuing to traverse for each drone group to obtain multiple groups of flight path combinations; wherein a group of flight path combinations includes a set of flight paths for each drone in the drone group to arrive at a corresponding single disaster-stricken area under the condition that each drone matches different single disaster-stricken areas, and comparing and determining the first drone group and the first flight route that meet the conditions that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized.

[0037] Among them, each UAV can complete the monitoring task of a single disaster-stricken area means that the UAV's endurance (battery power) supports its round-trip from its own location to the single disaster-stricken area and flight monitoring in the single disaster-stricken area. Flight monitoring in a single disaster-stricken area requires that the flight path in the single disaster-stricken area corresponds to a shooting range that covers the entire range of the single disaster-stricken area; the minimum total time to complete the monitoring task is the minimum sum of the flight path of each UAV to complete the monitoring task of a single disaster-stricken area divided by the preset flight speed.

[0038] According to the planned first flight route of each drone in the first drone group, the corresponding drone is called to go to the disaster-stricken area to perform the monitoring task, and the real-time data of the disaster-stricken area photographed during the monitoring process is obtained.

[0039] S5. Based on the real-time data of the disaster-stricken area, identify the distribution of key points in the disaster-stricken area and predict the spreading direction and scope of the disaster-stricken area.

[0040] Specifically, image analysis technology is used to analyze real-time data of the disaster-stricken area, and the distribution of key points in the disaster-stricken area is identified according to the predetermined rules for executing emergency tasks, that is, the distribution of key points in the emergency scene is identified; for example, if the determined scene is a fire scene, the key points such as fire source points, high temperature points and boundary points in the predetermined rules for executing emergency tasks are monitored and identified, including fire source points, high temperature points (points greater than a preset temperature threshold) and flame boundary points in the image.

[0041] At the same time, combined with the environmental data (wind direction and wind speed) in the alarm information, the real-time image data of the disaster-stricken area and the environmental data are input into the matching disaster prediction model to obtain the predicted image of the disaster-stricken area; the predicted image of the disaster-stricken area is compared with the real-time image of the disaster-stricken area to obtain the predicted spreading direction and range of the disaster-stricken area. The disaster prediction model includes multiple types, each corresponding to an emergency scenario. Each disaster prediction model uses a neural network model, and the input is the image and environmental data of the disaster-stricken area collected in real time, and the image of the disaster-stricken area in the future is output. It is generated by training multiple historical images of the disaster-stricken area and the corresponding historical environmental data. Among them, the future period can be set according to artificial settings, and the prediction can be set periodically, thereby providing a basis for the subsequent dynamic generation of rescue tasks.

[0042] S6. Determine the rescue area, divide the rescue area and generate rescue tasks for the rescue area, plan the paths of the drones in the drone swarm according to the rescue tasks in the rescue area and call the drone swarm to collaboratively perform the rescue tasks.

[0043] Specifically, the rescue area is determined based on the distribution of key points of the identified disaster-stricken area and the predicted spreading direction and range of the disaster-stricken area. That is, the rescue area range is the range of the predicted disaster-stricken area, which can be determined based on the boundary points and spreading direction and range of the disaster-stricken area.

[0044] The rescue areas are divided and determined, and a rescue mission is generated for each divided rescue area. The specific division can be based on the area of ​​the largest rescue area of ​​a single UAV in history as a division standard to further divide the rescue area.

[0045] Determine the number of drones required according to the number of rescue missions in the generated rescue area, and determine the type of drone required according to the type of drone matched by the determined emergency scenario; query the drone information around the rescue area according to the location of the obtained rescue area, combined with the recorded drone site or the received drone communication information, obtain the data information of the drones around the rescue area, and pre-query the drones that meet the drone type and are in an idle state and have sufficient power in combination with the data information of the drones around the rescue area, generate a second drone set, and select drones that meet the required number from the second drone set to construct a second drone group; use a multi-objective optimization algorithm to plan the second flight route of each drone in the second drone group, so that each drone can complete the rescue mission of a single rescue area and the total time to complete the rescue mission is minimized. Wherein, the process of determining the second drone group and planning the first path of each drone in the second drone group under the condition of meeting the restriction requirements includes: pre-traversing and forming multiple drone groups, and continuing to traverse for each drone group to obtain multiple groups of flight path combinations; wherein a group of flight path combinations includes the flight path set of each drone arriving at the corresponding single rescue area under the condition that each drone in the drone group matches different single rescue areas, and compares and determines the second drone group and the second flight route that meet the conditions that each drone can complete the rescue mission of a single disaster area and the total time to complete the rescue mission is minimized.

[0046] Among them, each UAV can complete the monitoring task of a single rescue area means that the UAV's endurance can support its round-trip from its own location to the location of a single rescue area and flight rescue in a single rescue area. Flight rescue in a single rescue area must meet the flight path of a single disaster-stricken area and the delivery of materials covers the entire range of a single rescue area; the minimum total time for completing the monitoring task is the minimum sum of the flight path of each UAV to complete the rescue task in a single rescue area divided by the preset flight speed.

[0047] According to the planned second flight route of each drone in the second drone group, the corresponding drone is called to go to the rescue area to perform the rescue mission.

[0048] By utilizing the above implementation methods, it is possible to dynamically generate drone task allocation and collaborative execution based on real-time monitoring content in the face of various emergency scenarios, thereby improving rescue efficiency and accuracy.

[0049] In a specific embodiment, considering different emergency scenarios and different disaster levels, the requirements for real-time monitoring and rescue of emergency scenario changes using drones are different. For example, in an earthquake emergency scenario, the requirements for monitoring accuracy and rescue accuracy for a severe disaster level are much greater than those for a general disaster level. Therefore, the disaster area and rescue area are divided and refined according to different disaster levels to achieve more detailed monitoring and rescue and improve rescue effectiveness. The method also includes: The disaster level corresponding to each disaster-stricken area is obtained according to the received alarm information; specifically, the alarm information includes the disaster level information of the disaster-stricken area, and each disaster-stricken area is determined in the recorded disaster areas, and then the level information of each disaster-stricken area is obtained accordingly; in this embodiment, different disaster levels in different emergency scenarios include: minor, general, and severe.

[0050] The divided disaster-stricken areas are further divided according to the disaster level corresponding to each divided disaster-stricken area; different disaster levels are preset with corresponding disaster area range division standards; the division standards for light, general and severe fire emergency scenes are based on the range of the first disaster-stricken area as a single disaster-stricken area division standard, the range of the second disaster-stricken area as a single disaster-stricken area division standard, and the range of the third disaster-stricken area as a single disaster-stricken area division standard, and the first disaster-stricken area, the second disaster-stricken area, and the third disaster-stricken area gradually decrease in order; the division standards for light, general and severe earthquake emergency scenes are based on the range of the fourth disaster-stricken area as a single disaster-stricken area division standard. The scope of the region is the standard for dividing a single disaster-affected area, the scope of the fifth disaster-affected area is the standard for dividing a single disaster-affected area, and the scope of the sixth disaster-affected area is the standard for dividing a single disaster-affected area. The fourth disaster-affected area, the fifth disaster-affected area, and the sixth disaster-affected area are gradually reduced in order; the standards for dividing chemical emergency scenes into minor, general, and severe ones are the seventh disaster-affected area, the eighth disaster-affected area, and the ninth disaster-affected area. The seventh disaster-affected area, the eighth disaster-affected area, and the ninth disaster-affected area are gradually reduced in order.

[0051] Input the key point distribution of the identified disaster-stricken area corresponding to each divided rescue area and the predicted spreading direction and range of the disaster-stricken area into the disaster level judgment model corresponding to the emergency scenario, and obtain the disaster level (minor, general, severe) corresponding to each rescue area; there are multiple disaster level judgment models, each disaster level judgment model matches an emergency scenario and adopts a neural network model, which is generated by training the key point distribution of the identified historical disaster-stricken area corresponding to the historical rescue area marked with the disaster level under the matched emergency scenario, and the predicted spreading direction and range of the historical disaster-stricken area; Each rescue area is further divided according to the disaster level output by each rescue area; wherein, different disaster levels are preset with corresponding rescue area range division standards. wherein, different disaster levels are preset with corresponding rescue area range division standards; the division standards for minor, general, and serious fire emergency scenes are respectively based on the scope of the first rescue area as the single rescue area division standard, the scope of the second rescue area as the single rescue area division standard, and the scope of the third rescue area as the single rescue area division standard, and the first rescue area, the second rescue area, and the third rescue area are gradually reduced in order; the division standards for minor, general, and serious earthquake emergency scenes are respectively based on the scope of the fourth rescue area as the single rescue area division standard, the scope of the fifth rescue area as the single rescue area division standard, and the scope of the sixth rescue area as the single rescue area division standard, and the fourth rescue area, the fifth rescue area, and the sixth rescue area are gradually reduced in order; the division standards for minor, general, and serious chemical emergency scenes are respectively based on the scope of the seventh rescue area as the single rescue area division standard, the scope of the eighth rescue area as the single rescue area division standard, and the scope of the ninth rescue area as the single rescue area division standard, and the seventh rescue area, the eighth rescue area, and the ninth rescue area are gradually reduced in order.

[0052] In a specific embodiment, in addition to using monitoring or rescue time as a path planning constraint, it is also possible to further comprehensively consider resource utilization constraints and select a path planning with a high resource utilization as much as possible, thereby improving rescue effectiveness while saving costs. The method includes: The power consumption data per kilometer obtained from the data information of drones around the disaster area; Considering the performance data of drones, based on the data information of drones around the disaster-stricken area and the matched drone types, similar to the above-mentioned embodiment, the required number of drones is determined according to the number of monitoring tasks for the generated disaster-stricken area, and the required type of drone is determined according to the drone type matched by the determined emergency scenario; combined with the data information of drones around the disaster-stricken area, drones that meet the drone type and are in an idle state and have sufficient power are pre-queried, a first drone set is generated, and drones that meet the required number are selected from the first drone set to reconstruct a third drone group; a third flight path of the third drone group is planned using a multi-objective optimization algorithm so that each drone can complete the monitoring task of a single disaster-stricken area and the weighted value of the total time to complete the monitoring task and the total power consumption is minimized; wherein, the weighted value of the total time to complete the monitoring task and the total power consumption is minimized as follows: the sum of the flight path of each drone corresponding to the monitoring task of a single disaster-stricken area divided by the preset flight speed and the sum of the power consumption required for each drone to complete the flight path of the monitoring task of a single disaster-stricken area is weighted and calculated to be the minimum; wherein the specific weight value can be set manually.

[0053] Invoking the corresponding drone to go to the disaster-stricken area to perform the monitoring task according to the third flight route of each drone in the planned third drone group replaces calling the corresponding drone to go to the disaster-stricken area to perform the monitoring task according to the first flight route of each drone in the planned first drone group.

[0054] Similarly, considering the performance data of drones, based on the data information of drones around the disaster area and the matched drone types, similar to the above-mentioned embodiment, the required number of drones is determined according to the number of rescue area monitoring tasks generated, and the required drone type is determined according to the determined drone type matching the emergency scenario; the data information of drones around the rescue area is obtained, and drones that meet the drone type and are in an idle state and have sufficient power are pre-queried, a second drone set is generated, and drones that meet the required number are selected from the second drone set to reconstruct a fourth drone group; a fourth flight route of each drone in the fourth drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the rescue mission of a single rescue area and the weighted value of the total time to complete the rescue mission and the total power consumption is minimized; According to the planned fourth flight route of each drone in the fourth drone group, the corresponding drone is called to go to the rescue area to perform the rescue mission instead of calling the corresponding drone to go to the disaster area to perform the rescue mission according to the planned second flight route of the second drone group.

[0055] In a specific embodiment, considering that there may be failures in the process of dispatching drones for monitoring and rescue, redundant drones are set to continue to complete monitoring and rescue, thereby further improving the effectiveness of rescue. The method also includes: The flight status data of the drones in the first drone group or the second drone group are received in real time, and the machine learning technology is used to predict the fault of the drone flight status data; if the prediction result is that a fault exists, a return instruction is generated for the drone with the fault.

[0056] At the same time, the drone will record the emergency mission execution process and upload it to the drone monitoring system in real time, including: the progress of emergency mission execution, including the type and area scope of completed monitoring or rescue missions; by obtaining the records of drones executing emergency missions, the type and area of ​​​​unexecuted emergency missions are determined according to the records of emergency missions executed by faulty drones, and recorded as the type of area to be supplemented and the area to be supplemented.

[0057] According to the type of emergency mission to be supplemented, a redundant drone (with sufficient power) of the corresponding type is selected from the original take-off station of the faulty drone, and a supplementary emergency mission route is generated from the current location of the redundant drone to the area where the emergency mission is to be supplemented, and the supplementary emergency mission route does not conflict with the first flight route or the second flight route of the original faulty drone.

[0058] In a specific embodiment, considering that due to emergencies, such as excessive battery life and increased battery consumption, the UAV that has completed the preset planned route according to the original support has insufficient power, in order to complete the assigned monitoring task or rescue task and reduce time consumption, the method further includes: The flight status data of each drone in the first drone group or the second drone group is received in real time, and it is determined whether the power of each drone reaches a minimum power preset threshold; wherein different models of drones have corresponding minimum power preset thresholds, and when determining whether the power of each drone reaches the minimum power preset threshold, a comparison is performed with the corresponding minimum power preset threshold according to the model of each drone.

[0059] If reached, a hovering command is sent to the UAV whose battery has reached the preset minimum battery threshold, and the energy supply UAV is called, so that the energy supply UAV arrives above the real-time position of the UAV that has reached the preset minimum battery threshold along the flight route of the UAV that has reached the preset minimum battery threshold, and after the energy supply UAV completes the replenishment of the UAV that has reached the preset minimum battery threshold, it returns to the flight starting point along the flight route of the UAV that has reached the preset minimum battery threshold.

[0060] In a specific embodiment, in order to further improve the monitoring and rescue efficiency in different emergency scenarios, a reasonable flight mode can be planned so that the drone can perform monitoring or rescue tasks more quickly and accurately. The method also includes: When the drone performs monitoring tasks in the disaster-stricken area or performs rescue tasks in the rescue area, the flight mode corresponding to the monitoring tasks in the disaster-stricken area or the rescue tasks in the rescue area is matched in advance according to the determined emergency scenarios; wherein each emergency scenario is matched with a preset flight mode for the drone to perform monitoring tasks and a flight mode for the drone to perform rescue tasks. Specifically, the preset flight mode for the drone to perform monitoring tasks in the fire emergency scenario uses the S-type flight mode, and the preset flight mode for the drone to perform rescue tasks uses the medium-high altitude flight mode to prevent burning; the preset flight mode for the drone to perform monitoring tasks in the earthquake emergency scenario uses the grid-type flight mode, and the preset flight mode for the drone to perform rescue tasks uses the low-altitude flight mode to achieve precise delivery; the preset flight mode for the drone to perform monitoring tasks in the chemical leak emergency scenario uses the S-type flight mode, and the preset flight mode for the drone to perform rescue tasks uses the low-altitude flight mode to achieve better neutralization and adsorption.

[0061] Complete monitoring missions in disaster-stricken areas or rescue missions in rescue areas according to the matching flight mode.

[0062] like Figure 2 As shown, the embodiment of the present application discloses a drone emergency management system, including: The disaster area and scene determination module 101 is used to receive alarm information and preliminarily determine the disaster area and emergency scene; the emergency scene includes: fire, earthquake and chemical leakage; obtain data information of drones around the disaster area; the data information includes location data and status data; The disaster emergency strategy matching module 102 is used to match the preset emergency strategy according to the emergency scenario, and each emergency scenario is provided with a preset emergency strategy matching it; the preset emergency strategy includes: predetermined rules for executing emergency tasks and types of drones for executing emergency tasks; wherein the predetermined rules for executing emergency tasks include key points of the executed emergency tasks; Execute the disaster-stricken area monitoring module 103, which is used to divide the disaster-stricken area and generate a number of disaster-stricken area monitoring tasks accordingly; based on the data information of the drones around the disaster-stricken area and the matched drone types, build a first drone group and use a multi-objective optimization algorithm to plan the first flight route of each drone in the first drone group, so that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized; according to the planned first flight route of each drone in the first drone group, call the corresponding drone to go to the disaster-stricken area to perform the monitoring task and obtain real-time data of the disaster-stricken area; use image analysis technology to analyze the real-time data of the disaster-stricken area, identify the distribution of key points in the disaster-stricken area according to the predetermined rules for executing emergency tasks, and predict the spread direction and range of the disaster-stricken area; The rescue module 104 is used to determine the rescue area according to the distribution of key points in the identified disaster-stricken area and the predicted spreading direction and range of the disaster-stricken area, divide the rescue area and generate a number of rescue tasks accordingly; obtain the drone data information around the rescue area, and build a second drone group based on the data information of the drones around the rescue area and the matched drone types, and use a multi-objective optimization algorithm to plan the second flight route of each drone in the second drone group, so that each drone can complete the rescue task of a single rescue area and the total time to complete the rescue task is minimized, and call the corresponding drone to the rescue area according to the planned second flight route of each drone in the second drone group to perform the rescue task.

[0063] The system further comprises: The module 105 for executing the emergency task to be supplemented is used to receive the flight status data of each drone in the first drone group or the second drone group in real time, and use machine learning technology to predict the fault of the drone's flight status data; if the prediction result is that there is a fault, a return instruction is generated for the drone with the fault; at the same time, the type and area of ​​the emergency task not executed are determined according to the record of executing the emergency task of the drone with the fault, and are recorded as the type of area for the emergency task to be supplemented and the area for the emergency task to be supplemented, respectively; according to the type of the emergency task to be supplemented, a redundant drone of the corresponding type is selected from the original take-off position station of the drone with the fault, and a supplementary emergency task route is generated from the current position of the redundant drone to the area for the emergency task to be supplemented, and the supplementary emergency task route does not conflict with the first flight route or the second flight route of the original drone with the fault.

[0064] The disaster-stricken area monitoring and rescue area rescue guarantee module 106 is used to receive the flight status data of each drone in the first drone group or the second drone group in real time, and determine whether the power of each drone has reached the minimum power preset threshold; if reached, a hovering command is sent to the drone whose power has reached the minimum power preset threshold, and an energy replenishment drone is called to make the energy replenishment drone arrive above the real-time position of the drone whose power has reached the minimum power preset threshold along the flight route of the drone whose power has reached the minimum power preset threshold, and after the energy replenishment drone completes the replenishment of the drone whose power has reached the minimum power preset threshold, it returns to the flight starting point along the flight route of the drone whose power has reached the minimum power preset threshold.

[0065] In a specific embodiment, the module 103 for executing disaster-stricken area monitoring is also used to match the flight mode corresponding to the monitoring task in the disaster-stricken area in advance according to the determined emergency scenario during the drone's execution of the monitoring task in the disaster-stricken area, and complete the monitoring task in the disaster-stricken area according to the matched flight mode; wherein each emergency scenario is matched with a preset drone flight mode for executing the monitoring task and a drone flight mode for executing the rescue task.

[0066] The rescue module 104 for executing rescue tasks in the rescue area is also used to match the corresponding flight mode for executing the rescue task in the rescue area in advance according to the determined emergency scenario during the process of the drone executing the rescue mission in the rescue area, and complete the rescue mission in the rescue area according to the matched flight mode; wherein each emergency scenario is matched with a preset flight mode for the drone to execute the monitoring task and a flight mode for the drone to execute the rescue mission.

[0067] In a specific embodiment, the disaster-affected area monitoring module 103 is further used to obtain the disaster level corresponding to each divided disaster-affected area according to the received alarm information; further divide the divided disaster-affected areas according to the obtained disaster level corresponding to each divided disaster-affected area; wherein different disaster levels are preset with corresponding disaster area range division standards; The rescue module 104 for executing rescue area rescue is further used for inputting the key point distribution of the identified disaster-stricken area corresponding to each divided rescue area and the predicted spreading direction and range of the disaster-stricken area into a disaster level judgment model that matches the corresponding emergency scenario, so as to obtain the disaster level corresponding to each rescue area; there are multiple disaster level judgment models, each disaster level judgment model matches an emergency scenario and adopts a neural network model, which is generated by training the key point distribution of the identified historical disaster-stricken area and the predicted spreading direction and range of the historical disaster-stricken area corresponding to the historical rescue area marked with the disaster level under the matched emergency scenario; each divided rescue area is further divided according to the disaster level division output by each divided rescue area; wherein, different disaster levels are preset with corresponding rescue area range division standards.

[0068] In a specific embodiment, the module 103 for executing the disaster-stricken area monitoring is also used to obtain data information of drones around the disaster-stricken area, which also includes drone performance data, and the performance data includes power consumption data per kilometer; considering the drone performance data, based on the data information of drones around the disaster-stricken area and the matching drone types, a third drone group is reconstructed and a third flight route of each drone in the third drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the monitoring task of a single disaster-stricken area and the weighted value of the total time to complete the monitoring task and the total power consumption is minimized; according to the planned third flight route of each drone in the third drone group, the corresponding drone is called to go to the disaster-stricken area to perform the monitoring task instead of calling the corresponding drone to go to the disaster-stricken area according to the planned first flight route of each drone in the first drone group to perform the monitoring task; The rescue module 104 for executing rescue in the rescue area is also used to obtain data information of drones around the disaster-stricken area, including drone performance data, and the performance data includes power consumption data per kilometer; considering the drone performance data, based on the data information of drones around the rescue area and the matching drone types, a fourth drone group is reconstructed and a fourth flight route of each drone in the fourth drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the rescue mission of a single rescue area and the weighted value of the total time to complete the rescue mission and the total power consumption is minimized; according to the planned fourth flight route of each drone in the fourth drone group, the corresponding drone is called to go to the rescue area to perform the rescue mission instead of calling the corresponding drone to go to the disaster-stricken area according to the planned second flight route of the second drone group.

[0069] The embodiment of the present application also discloses a computer-readable storage medium.

[0070] Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and executed such as the above-mentioned drone emergency management method. The computer-readable storage medium includes, for example: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0071] The embodiment of the present application also discloses a computer device.

[0072] Specifically, the computer device includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and execute the above-mentioned drone emergency management method.

[0073] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A drone emergency management method, characterized in that: include: Receive alarm information and preliminarily determine the disaster-affected areas and emergency scenarios; The emergency scenarios include: fire, earthquake and chemical leakage; obtaining data information of drones around the disaster area; the data information includes location data and status data; According to the emergency scenario matching preset emergency strategies, each emergency scenario is provided with a preset emergency strategy matching it; the preset emergency strategy includes: predetermined rules for executing emergency tasks, and types of drones for executing emergency tasks; wherein the predetermined rules for executing emergency tasks include key points of the executed emergency tasks; Divide the disaster-stricken area and generate a number of disaster-stricken area monitoring tasks accordingly; construct a first drone group based on the data information of drones around the disaster-stricken area and the matched drone types, and use a multi-objective optimization algorithm to plan the first flight route of each drone in the first drone group, so that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized; call the corresponding drone to the disaster-stricken area according to the planned first flight route of each drone in the first drone group to perform the monitoring task and obtain real-time data of the disaster-stricken area; use image analysis technology to analyze the real-time data of the disaster-stricken area, identify the distribution of key points in the disaster-stricken area according to the predetermined rules for executing emergency tasks, and predict the spread direction and range of the disaster-stricken area; The rescue area is determined based on the distribution of key points in the identified disaster-stricken area and the predicted spreading direction and range of the disaster-stricken area, the rescue area is divided and a number of rescue tasks are generated accordingly; the drone data information around the rescue area is obtained, and based on the data information of the drones around the rescue area and the matched drone types, a second drone swarm is constructed and a second flight route for each drone in the second drone swarm is planned using a multi-objective optimization algorithm, so that each drone can complete the rescue task of a single rescue area and the total time to complete the rescue task is minimized; and the corresponding drone is called to the rescue area to perform the rescue task according to the planned second flight route of each drone in the second drone swarm.

2. The drone emergency management method according to claim 1, characterized in that: Also includes: Obtaining the disaster level corresponding to each divided disaster area according to the received alarm information; further dividing the divided disaster area according to the obtained disaster level corresponding to each divided disaster area; wherein different disaster levels are preset with corresponding disaster area range division standards; The key point distribution of the identified disaster-stricken areas corresponding to each divided rescue area and the predicted spreading direction and range of the disaster-stricken areas are input into the disaster level judgment model corresponding to the emergency scenario to obtain the disaster level corresponding to each rescue area; there are multiple disaster level judgment models, each disaster level judgment model matches an emergency scenario and adopts a neural network model, which is generated by training the key point distribution of the identified historical disaster-stricken areas and the predicted spreading direction and range of the historical disaster-stricken areas corresponding to the historical rescue areas marked with disaster levels under the matched emergency scenarios; each divided rescue area is further divided according to the disaster level division output by each divided rescue area; wherein different disaster levels are preset with corresponding rescue area range division standards.

3. The drone emergency management method according to claim 1, characterized in that: Also includes: The acquired data information of drones around the disaster-stricken area also includes drone performance data, and the performance data includes power consumption data per kilometer; Considering the performance data of the drones, based on the data information of the drones around the disaster-stricken area and the matched drone types, the third drone group is reconstructed and the third flight route of each drone in the third drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the monitoring task of a single disaster-stricken area and the weighted value of the total time to complete the monitoring task and the total power consumption is minimized; the corresponding drone is called to go to the disaster-stricken area to perform the monitoring task according to the planned third flight route of each drone in the third drone group, instead of calling the corresponding drone to go to the disaster-stricken area to perform the monitoring task according to the planned first flight route of each drone in the first drone group; Taking into account the drone performance data, based on the data information of the drones around the rescue area and the matching drone types, the fourth drone swarm is reconstructed and the fourth flight route of each drone in the fourth drone swarm is planned using a multi-objective optimization algorithm, so that each drone can complete the rescue mission in a single rescue area and the weighted values ​​of the total time to complete the rescue mission and the total power consumption are minimized; according to the planned fourth flight route of each drone in the fourth drone swarm, the corresponding drone is called to go to the rescue area to perform the rescue mission instead of calling the corresponding drone to go to the disaster-stricken area to perform the rescue mission according to the planned second flight route of the second drone swarm.

4. The drone emergency management method according to claim 1, characterized in that: Also includes: receiving flight status data of each drone in the first drone group or the second drone group in real time, and using machine learning technology to perform fault prediction on the flight status data of the drone; If the prediction result is that there is a fault, a return instruction is generated for the faulty UAV; at the same time, the type and area of ​​the emergency task that was not performed are determined based on the record of the emergency task performed by the faulty UAV, and recorded as the type of emergency task area to be supplemented and the area to be supplemented; According to the type of emergency mission to be supplemented, a redundant drone of the corresponding type is selected from the original take-off position station of the faulty drone, and a supplementary emergency mission route is generated from the current location of the redundant drone to the area where the emergency mission is to be supplemented, and the supplementary emergency mission route does not conflict with the first flight route or the second flight route of the original faulty drone.

5. The drone emergency management method according to claim 1, characterized in that: Also includes: receiving the flight status data of each drone in the first drone group or the second drone group in real time, and determining whether the power of each drone reaches a preset minimum power threshold; If reached, a hovering command is sent to the UAV whose battery has reached the preset minimum battery threshold, and the energy supply UAV is called, so that the energy supply UAV arrives above the real-time position of the UAV that has reached the preset minimum battery threshold along the flight route of the UAV that has reached the preset minimum battery threshold, and after the energy supply UAV completes the replenishment of the UAV that has reached the preset minimum battery threshold, it returns to the flight starting point along the flight route of the UAV that has reached the preset minimum battery threshold.

6. The drone emergency management method according to claim 2, characterized in that: Also includes: When the UAV performs monitoring tasks in the disaster-stricken area or performs rescue tasks in the rescue area, the flight mode corresponding to the monitoring tasks in the disaster-stricken area or the rescue tasks in the rescue area is matched in advance according to the determined emergency scenario, and the monitoring tasks in the disaster-stricken area or the rescue tasks in the rescue area are completed according to the matched flight mode; wherein each emergency scenario is matched with a preset flight mode for the UAV to perform monitoring tasks and a flight mode for the UAV to perform rescue tasks.

7. The drone emergency management method according to claim 1, characterized in that: The types of drones that perform emergency tasks in the preset emergency strategy according to the emergency scenario matching include: The types of drones that perform emergency tasks in the preset emergency strategies that match fire emergency scenarios include thermal imaging drones, high-definition camera drones, and drones equipped with fire-fighting materials or fire-fighting devices; the types of drones that perform emergency tasks in the preset emergency strategies that match earthquake emergency scenarios include infrared detection drones, high-definition camera drones, and drones equipped with rescue supplies; the types of drones that perform emergency tasks in the preset emergency strategies that match chemical leak emergency scenarios include drones equipped with chemical sensors, high-definition camera drones, and drones equipped with chemical neutralization materials or chemical adsorption materials.

8. A drone emergency management system, characterized in that: include: A disaster-affected area and scene determination module, used to receive alarm information and preliminarily determine the disaster-affected area and emergency scene; The emergency scenarios include: fire, earthquake and chemical leakage; obtaining data information of drones around the disaster area; the data information includes location data and status data; The disaster emergency strategy matching module is used to match the preset emergency strategy according to the emergency scenario. Each emergency scenario is provided with a preset emergency strategy matching it; the preset emergency strategy includes: predetermined rules for executing emergency tasks and types of drones for executing emergency tasks; wherein the predetermined rules for executing emergency tasks include key points of the executed emergency tasks; A disaster-stricken area monitoring module is executed to divide the disaster-stricken area and generate a number of disaster-stricken area monitoring tasks accordingly; based on the data information of the drones around the disaster-stricken area and the matched drone types, a first drone group is constructed and a first flight route of each drone in the first drone group is planned using a multi-objective optimization algorithm, so that each drone can complete the monitoring task of a single disaster-stricken area and the total time to complete the monitoring task is minimized; according to the planned first flight route of each drone in the first drone group, the corresponding drone is called to go to the disaster-stricken area to perform the monitoring task and obtain real-time data of the disaster-stricken area; the real-time data of the disaster-stricken area is analyzed using image analysis technology, and the distribution of key points in the disaster-stricken area is identified according to the predetermined rules for executing emergency tasks, and the spreading direction and range of the disaster-stricken area are predicted; The rescue module of the rescue area is executed, which is used to determine the rescue area according to the distribution of key points in the identified disaster-stricken area and the predicted spreading direction and range of the disaster-stricken area, divide the rescue area and generate a number of rescue tasks accordingly; obtain the drone data information around the rescue area, and build a second drone group based on the data information of the drones around the rescue area and the matching drone types, and use a multi-objective optimization algorithm to plan the second flight route of each drone in the second drone group, so that each drone can complete the rescue task of a single rescue area and the total time to complete the rescue task is minimized, and call the corresponding drone to the rescue area according to the planned second flight route of each drone in the second drone group to perform the rescue task.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. A computer device, characterized in that: The computer device comprises a memory, a processor and a program stored and executable on the memory, and the program implements the steps of the method according to any one of claims 1 to 7 when executed by the processor.

Images