Forest fire prevention unmanned aerial vehicle patrol early warning system and method based on edge calculation

Abstract

The invention provides a forest fire prevention unmanned aerial vehicle patrol early warning system and method based on edge calculation, and relates to the technical field of unmanned aerial vehicle patrol early warning. The patrol early warning system comprises an unmanned aerial vehicle platform, a data acquisition module, a data processing and analysis module, a communication and early warning module and a ground control and scheduling platform. According to the invention, calculation and data storage are close to a data source through edge calculation, the data collected by the unmanned aerial vehicle is processed in real time at an edge calculation node without being transmitted to a remote cloud and returning a result, the data processing and response time is greatly shortened, fire hazards can be found in seconds or even shorter time, and early warning can be given out. And the system can analyze data in real time in the flight process of the unmanned aerial vehicle, provides real-time decision support for a ground control center, and can quickly formulate a fire-fighting scheme according to information such as fire behavior and wind direction when a fire occurs, so that the fire-fighting efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of UAV patrol warning technology, and specifically to a forest fire prevention UAV patrol warning system and method based on edge computing. Background Art

[0002] Various natural disasters are often encountered in the process of forest reproduction, and the harm caused by forest fires is extremely serious. Forest fires refer to forest fires that lose human control, spread and expand freely in the forest, and cause certain harm and losses to the forest, forest ecosystem and humans. Forest fires are a natural disaster that is highly sudden, destructive, and difficult to handle and rescue.

[0003] my country has a forest area of ​​about 2.08×107hm 2 The forest volume is 1.5137×109m 3 , the forest coverage rate is 21.63%. Frequent forest fires have caused huge losses to my country's precious forest resources. Forest fires are often difficult to extinguish, requiring huge manpower and funds, and are extremely likely to cause casualties. It also takes decades or even hundreds of years for the ecological environment to recover after the fire.

[0004] At present, the main forest fire monitoring methods in my country include satellite monitoring, observation by watchtowers, manned aircraft observation, and manual inspections. The above methods have the following disadvantages:

[0005] 1. Satellite monitoring has high requirements on climate and can only take pictures and images with low resolution;

[0006] 2. The observation from the lookout tower cannot accurately locate the fire point, which may delay the best time to put out the fire;

[0007] 3. The cost of using manned aircraft for aerial forest protection is high and it is a high-risk industry. The operation and operation of manned aircraft are complicated, and safety accidents are prone to occur during patrol observation, fire handling, and air landing;

[0008] 4. Manual inspections are greatly affected by terrain factors, are prone to danger, have a small coverage area, and are inefficient.

[0009] Therefore, those skilled in the art provide a forest fire prevention drone patrol warning system and method based on edge computing to solve the problems raised in the above background technology. Summary of the invention

[0010] 1. Technical issues to be resolved

[0011] In view of the shortcomings of the prior art, the present invention provides a forest fire prevention drone patrol warning system and method based on edge computing to solve the problem.

[0012] (II) Technical solution

[0013] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0014] A forest fire prevention drone patrol warning system based on edge computing, including a drone platform, a data acquisition module, a data processing and analysis module, a communication and warning module, and a ground control and dispatching platform;

[0015] UAV platform, using multi-rotor UAV or fixed-wing UAV, select the appropriate model according to the needs of the task, use fixed-wing UAV to patrol a large area for a long time, once a suspected fire area is found, send multi-rotor UAV to conduct a low-altitude detailed inspection of the suspected fire area to determine whether the suspected fire area is a fire;

[0016] The data acquisition module is used to collect forest images, videos and meteorological data in real time while the drone is flying according to the preset mission. The collected data is transmitted to the data processing and analysis module through a wireless communication link. In order to ensure the stability and efficiency of data transmission, multi-link redundant transmission technology can be used;

[0017] The data processing and analysis module performs real-time analysis on various sensor data, detects fire sources through thermal imaging technology, and detects abnormal smoke concentrations through gas sensors. The edge computing device can complete data processing locally on the drone and make preliminary judgments, reducing the delay of relying on remote server processing;

[0018] Communication and warning module: Through 5G, LoRa, and satellite communication technologies, the drone uploads real-time data and processing results to the ground command center. When signs of fire are found, the warning mechanism is automatically triggered to send an alarm to relevant departments through SMS, phone calls, and emails. The command center can receive fire alarm information and respond and dispatch in real time.

[0019] The ground control and dispatch platform is used to monitor the flight status of the drone and the data it collects in real time. The dispatch platform can display the drone's flight path, sensor data, and fire warning information.

[0020] Further, the UAV platform includes a sensor unit, an edge computing device, and a flight control system;

[0021] Sensor units, mainly including thermal imaging infrared sensors, high-definition visible light cameras, gas sensors (such as smoke sensors), wind speed and direction sensors, which can monitor temperature changes, smoke concentrations and other fire-related characteristics in real time;

[0022] Edge computing devices are equipped on drones to process data from sensors in real time for preliminary identification, analysis and early warning of fires. Edge computing can greatly reduce data transmission delays and improve response speed.

[0023] The flight control system is used to realize the UAV's autonomous flight, route planning, automatic return and other functions, ensuring that the UAV flies accurately according to the preset patrol path.

[0024] Furthermore, the data acquisition module specifically includes a high-definition camera, an infrared thermal imager, and a meteorological sensor;

[0025] High-definition cameras can capture high-resolution visible light images to clearly capture forest vegetation conditions, topography, and possible fire sources and smoke anomalies;

[0026] Infrared thermal imagers can detect the thermal radiation of objects in the forest and effectively identify smoldering fire sources or areas with abnormally high temperatures hidden in the grass and trees. The detection accuracy can reach 0.1°C and can accurately locate the fire source at night or in a dense smoke environment.

[0027] Meteorological sensors collect meteorological data such as wind speed, wind direction, humidity, and temperature. These data are very important for fire risk assessment and fire spread prediction.

[0028] Furthermore, the data processing and analysis module specifically includes an edge computing node, a fire prediction unit, and a data fusion and intelligent decision-making unit;

[0029] Edge computing nodes, after receiving the data transmitted by the drone, the data processing and analysis module immediately processes and analyzes the data. For image and video data, the pre-trained deep learning model is used for target detection to identify possible fire sources and smoke abnormal targets;

[0030] Fire prediction unit, based on historical data and meteorological data, combined with machine learning or deep learning algorithms, predicts fire risks and provides decision support for fire prevention work;

[0031] Specifically, it includes: using infrared thermal imaging data to identify areas where the temperature suddenly rises, which may indicate a fire source; using gas sensor data to analyze sudden changes in smoke concentration, usually combined with environmental parameters (such as wind speed) for comprehensive evaluation; analyzing images captured by high-definition cameras, and applying computer vision technology (such as target detection and image segmentation) to identify signs of fire such as smoke and flames; predicting the spread of fire based on meteorological sensor data (such as wind speed, humidity, temperature, etc.), combined with historical fire data, and using edge computing to predict the possible spread direction and rate of fire;

[0032] The data fusion and intelligent decision-making unit integrates data from different sensors to improve the accuracy of fire identification. Using big data analysis and pattern recognition technology, the system can accurately predict potential fire risk areas.

[0033] Furthermore, the edge computing node specifically includes a data processing unit, a data storage unit, and a communication module;

[0034] The data processing unit is equipped with a high-performance processor to process the images, videos and meteorological data collected by the drone in real time;

[0035] Data storage unit, used to store a certain amount of historical data and temporary data, including recent patrol images, meteorological data and processing results, so that local data query and analysis can still be performed when the network connection is unstable, and it can also provide data support for subsequent data analysis and model optimization;

[0036] The communication module is used to realize two-way communication between the UAV and the ground control center, and adopts high-speed and stable wireless communication technology, such as 5G or dedicated microwave communication frequency band, to ensure high-speed data transmission.

[0037] Furthermore, the communication and warning module specifically includes a wireless communication unit and a warning and response unit;

[0038] Wireless communication unit, which realizes two-way communication between the UAV and the ground control center. The UAV uploads real-time data and processing results to the ground command center;

[0039] The early warning and response unit, when the edge computing node detects a fire hazard or determines that a fire has occurred, it immediately generates an early warning message and sends it to the ground control center. After receiving the early warning message, the early warning release system of the ground control center issues an alarm in accordance with the preset early warning method and activates the emergency response mechanism.

[0040] Furthermore, the ground control and dispatching platform specifically includes a monitoring terminal, a task dispatching and planning unit, and a data management unit;

[0041] Monitoring terminal: staff can view the flight status of the drone (including position, altitude, speed, etc.), data collection, and processing results and warning information of the edge computing node in real time through the monitoring terminal;

[0042] Task scheduling and planning unit: Based on monitoring data, the command center can schedule tasks for drones, such as adjusting flight paths and increasing patrol frequency;

[0043] The data management unit manages the massive data generated during the drone patrol process, including data storage, backup, retrieval and analysis. It can mine and analyze historical data, summarize the patterns and characteristics of forest fires, and provide a basis for the formulation of forest fire prevention strategies. The system can also store patrol data and fire warning information in the cloud for subsequent analysis and research.

[0044] Furthermore, a forest fire prevention drone patrol warning method based on edge computing includes the following steps:

[0045] Step S1. According to the area, topography and key fire prevention areas of the forest area, a patrol mission plan for the drone is formulated at the ground control center;

[0046] Step S2. After the drone takes off, it flies to the designated location and starts collecting data. The sensors on the drone collect data in real time. The thermal imaging sensor monitors temperature anomalies, the gas sensor monitors smoke concentration, and the wind speed and direction sensor monitors fire propagation conditions.

[0047] Step S3. The edge computing device on the drone processes the collected data in real time, uses a preset fire recognition algorithm to perform preliminary analysis to determine whether there is a fire hazard. For image and video data, a pre-trained deep learning model is used for target detection to identify possible fire sources and smoke abnormal targets;

[0048] Step S4. Once a potential fire is discovered, the edge computing node immediately issues an alarm and transmits the alarm information to the ground command center through the wireless communication module. The ground command center responds to the fire risk according to the emergency response.

[0049] Step S5. After the emergency response is completed, the drone continues its patrol mission and uploads real-time data to the ground control platform for remote monitoring and management;

[0050] Step S6. After each patrol mission, the ground control center conducts a comprehensive analysis of the data collected by the drone and the processing results of the edge computing node;

[0051] Step S7. Optimize the deep learning model based on the analysis results. At the same time, update basic information such as map data, vegetation data, and meteorological data of the forest area to provide a more accurate basis for the next patrol mission.

[0052] Furthermore, formulating the patrol mission plan of the UAV in step S1 includes determining the take-off point, cruising route, flight altitude, speed and data collection frequency parameters of the UAV, and sending these mission information to the flight control system of the UAV.

[0053] Furthermore, after the drone takes off in step S2, it flies to the designated location and starts collecting data, including first using a fixed-wing drone to conduct a long-term patrol of a large-scale patrol area. Once an area suspected of fire is found, a multi-rotor drone is dispatched to conduct a low-altitude detailed patrol of the suspected fire area.

[0054] (III) Beneficial effects

[0055] The present invention provides a forest fire prevention UAV patrol warning system and method based on edge computing.

[0056] It has the following beneficial effects:

[0057] 1. The present invention provides a forest fire prevention drone patrol warning system and method based on edge computing. Through edge computing, computing and data storage are close to the data source. The data collected by the drone is processed instantly at the edge computing node without the need to be transmitted to the remote cloud and then the results are returned. This greatly shortens the data processing and response time, and can detect fire hazards and issue warnings in seconds or even shorter, so that staff can respond quickly and gain more time for firefighting. The system can analyze data in real time during the flight of the drone and provide real-time decision support for the ground control center. When a fire occurs, it can quickly formulate a firefighting plan based on information such as the fire intensity and wind direction to improve firefighting efficiency.

[0058] 2. The present invention provides a forest fire prevention drone patrol warning system and method based on edge computing. The edge computing nodes are equipped with high-performance processors and dedicated chips, which can quickly process a large amount of images, videos and meteorological data collected by drones, reduce the burden of cloud computing, and improve data processing efficiency. It can simultaneously process multiple video streams and a large amount of sensor data to achieve all-round and multi-dimensional monitoring of forests. In addition, with the help of intelligent algorithms such as deep learning and machine learning, the edge computing nodes can deeply mine and analyze the data, automatically identify signs of fire such as fire sources, smoke, and abnormally high temperature areas, and can also evaluate and predict fire risks, thereby improving the accuracy and reliability of early warnings.

[0059] 3. The present invention provides a forest fire prevention drone patrol and early warning system and method based on edge computing. The drone is used for patrol. The drone has the characteristics of flexibility, high flight speed and wide coverage. It can quickly reach remote mountainous areas and dense forest areas that are difficult to reach manually for patrol, effectively expanding the monitoring range of forest fire prevention and improving patrol efficiency. The system can realize automatic take-off, cruising and landing of the drone, support 24-hour uninterrupted patrol, and make up for the time and space limitations of manual patrol. Especially at night and in severe weather conditions, the drone can still perform patrol tasks normally, greatly improving the monitoring capability of forest fires. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a schematic diagram of the structure of the forest fire prevention drone patrol warning system based on edge computing of the present invention;

[0061] Figure 2 This is a schematic diagram of the structure of the UAV platform of the present invention;

[0062] Figure 3 It is a schematic diagram of the structure of the data acquisition module of the present invention;

[0063] Figure 4 It is a schematic diagram of the structure of the data processing and analysis module of the present invention;

[0064] Figure 5 This is a schematic diagram of the edge computing node structure of the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the specific embodiments of the present invention to clearly and completely describe the technical solutions in the specific embodiments of the present invention. Obviously, the specific embodiments described are only part of the specific embodiments of the present invention, not all of the specific embodiments. Based on the specific embodiments of the present invention, all other specific embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] like Figure 1-5 As shown, a specific embodiment of the present invention provides a forest fire prevention drone patrol warning system based on edge computing, including a drone platform, a data acquisition module, a data processing and analysis module, a communication and warning module, and a ground control and scheduling platform;

[0068] UAV platform, using multi-rotor UAV or fixed-wing UAV, select the appropriate model according to the needs of the task, use fixed-wing UAV to patrol a large area for a long time, once a suspected fire area is found, send multi-rotor UAV to conduct a low-altitude detailed inspection of the suspected fire area to determine whether the suspected fire area is a fire;

[0069] The data acquisition module is used to collect forest images, videos and meteorological data in real time while the drone is flying according to the preset mission. The collected data is transmitted to the data processing and analysis module through a wireless communication link. In order to ensure the stability and efficiency of data transmission, multi-link redundant transmission technology can be used;

[0070] The data processing and analysis module performs real-time analysis on various sensor data, detects fire sources through thermal imaging technology, and detects abnormal smoke concentrations through gas sensors. The edge computing device can complete data processing locally on the drone and make preliminary judgments, reducing the delay of relying on remote server processing;

[0071] Communication and warning module: Through 5G, LoRa, and satellite communication technologies, the drone uploads real-time data and processing results to the ground command center. When signs of fire are found, the warning mechanism is automatically triggered to send an alarm to relevant departments through SMS, phone calls, and emails. The command center can receive fire alarm information and respond and dispatch in real time.

[0072] The ground control and dispatch platform is used to monitor the flight status of the drone and the data it collects in real time. The dispatch platform can display the drone's flight path, sensor data, and fire warning information.

[0073] The drone platform includes a sensor unit, an edge computing device, and a flight control system;

[0074] Sensor units, mainly including thermal imaging infrared sensors, high-definition visible light cameras, gas sensors (such as smoke sensors), wind speed and direction sensors, which can monitor temperature changes, smoke concentrations and other fire-related characteristics in real time;

[0075] Edge computing devices are equipped on drones to process data from sensors in real time for preliminary identification, analysis and early warning of fires. Edge computing can greatly reduce data transmission delays and improve response speed.

[0076] The flight control system is used to realize the UAV's autonomous flight, route planning, automatic return and other functions, ensuring that the UAV flies accurately according to the preset patrol path.

[0077] The data acquisition module specifically includes a high-definition camera, an infrared thermal imager, and a meteorological sensor;

[0078] High-definition cameras can capture high-resolution visible light images to clearly capture forest vegetation conditions, topography, and possible fire sources and smoke anomalies. For example, the camera resolution can reach 4K or even higher, with optical zoom function, so that small signs of fire sources can be found at a long distance;

[0079] Infrared thermal imagers can detect the thermal radiation of objects in the forest and effectively identify smoldering fire sources or areas with abnormally high temperatures hidden in the grass and trees. The detection accuracy can reach 0.1°C and can accurately locate the fire source at night or in a dense smoke environment.

[0080] Meteorological sensors collect meteorological data such as wind speed, wind direction, humidity, and temperature. These data are very important for fire risk assessment and fire spread prediction; for example, the wind speed sensor can accurately measure the wind speed range of 0-50m / s, and the humidity sensor can measure the accuracy of ±2%RH.

[0081] The data processing and analysis module specifically includes an edge computing node, a fire prediction unit, and a data fusion and intelligent decision-making unit;

[0082] Edge computing nodes, after receiving the data transmitted by the drone, the data processing and analysis module immediately processes and analyzes the data. For image and video data, the pre-trained deep learning model is used for target detection to identify possible fire sources and smoke abnormal targets;

[0083] Fire prediction unit, based on historical data and meteorological data, combined with machine learning or deep learning algorithms, predicts fire risks and provides decision support for fire prevention work;

[0084] Specifically, it includes: using infrared thermal imaging data to identify areas where the temperature suddenly rises, which may indicate a fire source; using gas sensor data to analyze sudden changes in smoke concentration, usually combined with environmental parameters (such as wind speed) for comprehensive evaluation; analyzing images captured by high-definition cameras, and applying computer vision technology (such as target detection and image segmentation) to identify signs of fire such as smoke and flames; predicting the spread of fire based on meteorological sensor data (such as wind speed, humidity, temperature, etc.), combined with historical fire data, and using edge computing to predict the possible spread direction and rate of fire;

[0085] The data fusion and intelligent decision-making unit integrates data from different sensors to improve the accuracy of fire identification. Using big data analysis and pattern recognition technology, the system can accurately predict potential fire risk areas.

[0086] The edge computing node specifically includes a data processing unit, a data storage unit, and a communication module;

[0087] The data processing unit is equipped with a high-performance processor to process the images, videos and meteorological data collected by the drone in real time. For example, it uses deep learning algorithms to analyze image and video data to quickly identify whether there are fire hazards, such as open flames and smoking areas. Compared with traditional cloud-based processing methods, it greatly reduces data transmission delays and achieves a response in seconds.

[0088] The data storage unit is used to store a certain amount of historical data and temporary data, including recent patrol images, meteorological data, and processing results, so that local data query and analysis can still be performed when the network connection is unstable. It can also provide data support for subsequent data analysis and model optimization. The storage capacity can be set according to actual needs, such as 500GB-1TB of storage space.

[0089] The communication module is used to realize two-way communication between the UAV and the ground control center. It adopts high-speed and stable wireless communication technology, such as 5G or dedicated microwave communication frequency band, to ensure high-speed data transmission. On the one hand, it receives the data collected by the UAV and transmits it to the data processing unit, and on the other hand, it sends the processing results and warning information to the ground control center and receives the command information from the ground control center. The transmission rate can reach hundreds of megabits per second.

[0090] The communication and warning module specifically includes a wireless communication unit and a warning and response unit;

[0091] Wireless communication unit, which realizes two-way communication between the UAV and the ground control center. The UAV uploads real-time data and processing results to the ground command center;

[0092] Early warning and response unit: When the edge computing node detects a fire hazard or determines that a fire has occurred, it immediately generates an early warning message and sends it to the ground control center. After receiving the early warning message, the early warning release system of the ground control center issues an alarm in accordance with the preset early warning method and activates the emergency response mechanism. For example, relevant staff quickly organize forest fire fighting forces based on the early warning information, determine the fire fighting plan, such as deploying fire trucks, helicopters and other equipment, and arrange firefighters to go to the fire scene to fight the fire. At the same time, according to the fire spread prediction results, surrounding residents are evacuated in advance and isolation zones are set up to minimize fire losses.

[0093] The ground control and dispatching platform specifically includes a monitoring terminal, a task dispatching and planning unit, and a data management unit;

[0094] Monitoring terminal, through which the staff can view the flight status of the drone (including position, altitude, speed, etc.), data collection, processing results of edge computing nodes and early warning information in real time; the monitoring terminal usually uses a large-screen display to intuitively display the map of the forest area and the patrol track of the drone, and can display images, videos and data analysis results in real time, such as displaying the forest temperature distribution in the form of a heat map;

[0095] Task scheduling and planning unit: Based on monitoring data, the command center can schedule tasks for drones, such as adjusting flight paths and increasing patrol frequency;

[0096] The data management unit manages the massive data generated during the drone patrol process, including data storage, backup, retrieval and analysis. It can mine and analyze historical data, summarize the patterns and characteristics of forest fires, and provide a basis for the formulation of forest fire prevention strategies. The system can also store patrol data and fire warning information in the cloud for subsequent analysis and research. For example, by correlating many years of meteorological data and fire occurrence data, the probability of fire under specific meteorological conditions can be found.

[0097] The forest fire prevention drone patrol warning method based on edge computing includes the following steps:

[0098] Step S1. According to the area, topography and key fire prevention areas of the forest area, a patrol mission plan for the drone is formulated at the ground control center;

[0099] Step S2. After the drone takes off, it flies to the designated location and starts collecting data. The sensors on the drone collect data in real time. The thermal imaging sensor monitors temperature anomalies, the gas sensor monitors smoke concentration, and the wind speed and direction sensor monitors fire propagation conditions.

[0100] Step S3. The edge computing device on the drone processes the collected data in real time, uses a preset fire recognition algorithm to perform preliminary analysis to determine whether there is a fire hazard. For image and video data, a pre-trained deep learning model is used for target detection to identify possible fire sources and smoke abnormal targets;

[0101] Step S4. Once a potential fire is discovered, the edge computing node immediately issues an alarm and transmits the alarm information to the ground command center through the wireless communication module. The ground command center responds to the fire risk according to the emergency response.

[0102] Step S5. After the emergency response is completed, the drone continues its patrol mission and uploads real-time data to the ground control platform for remote monitoring and management;

[0103] Step S6. After each patrol mission, the ground control center conducts a comprehensive analysis of the data collected by the drone and the processing results of the edge computing node;

[0104] Step S7. Optimize the deep learning model based on the analysis results. At the same time, update basic information such as map data, vegetation data, and meteorological data of the forest area to provide a more accurate basis for the next patrol mission.

[0105] Formulating the patrol mission plan of the UAV in step S1 includes determining the take-off point, cruising route, flight altitude, speed and data collection frequency parameters of the UAV, and sending these mission information to the flight control system of the UAV.

[0106] After the drone takes off in step S2, it flies to the designated location and starts collecting data, including first using a fixed-wing drone to patrol a large area for a long time. Once an area suspected of fire is found, a multi-rotor drone is dispatched to conduct a low-altitude detailed patrol of the suspected fire area.

[0107] The present invention adopts the above technical solution, and uses edge computing to store calculations and data close to the data source. The data collected by the drone is processed instantly at the edge computing node, without the need to transmit it to the remote cloud and then return the results. This greatly shortens the data processing and response time, and can detect fire hazards and issue early warnings in seconds or even shorter, allowing staff to respond quickly and gain more time for firefighting. The system can analyze data in real time during the flight of the drone and provide real-time decision support for the ground control center. When a fire occurs, it can quickly formulate a firefighting plan based on information such as the fire intensity and wind direction, thereby improving firefighting efficiency.

[0108] The edge computing nodes of the present invention are equipped with high-performance processors and dedicated chips, which can quickly process large amounts of images, videos and meteorological data collected by drones, reduce the burden of cloud computing, improve data processing efficiency, and can simultaneously process multiple video streams and a large amount of sensor data to achieve all-round and multi-dimensional monitoring of forests. In addition, with the help of intelligent algorithms such as deep learning and machine learning, edge computing nodes can deeply mine and analyze data, automatically identify signs of fire such as fire sources, smoke, and abnormally high temperature areas, and can also evaluate and predict fire risks, thereby improving the accuracy and reliability of early warnings.

[0109] The present invention uses drones for patrolling. Drones are flexible, fast in flight speed, and have a wide coverage. They can quickly reach remote mountainous areas and dense forest areas that are difficult to reach manually for patrolling, effectively expanding the monitoring range of forest fire prevention and improving patrol efficiency. The system can realize automatic takeoff, cruising, and landing of drones, and supports 24-hour uninterrupted patrolling, making up for the time and space limitations of manual patrolling. Especially at night and under severe weather conditions, drones can still perform patrol tasks normally, greatly improving the monitoring capabilities of forest fires.

[0110] Although specific embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the specific embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A forest fire prevention drone patrol warning system based on edge computing, characterized in that: It includes UAV platform, data acquisition module, data processing and analysis module, communication and early warning module, and ground control and dispatch platform; UAV platform, using multi-rotor UAV or fixed-wing UAV, select the appropriate model according to the needs of the task, use fixed-wing UAV to patrol a large area for a long time, once a suspected fire area is found, send multi-rotor UAV to conduct a low-altitude detailed inspection of the suspected fire area to determine whether the suspected fire area is a fire; The data acquisition module is used to collect forest images, videos and meteorological data in real time while the drone is flying according to the preset mission. The collected data is transmitted to the data processing and analysis module through a wireless communication link. In order to ensure the stability and efficiency of data transmission, multi-link redundant transmission technology can be used; The data processing and analysis module performs real-time analysis on various sensor data, detects fire sources through thermal imaging technology, and detects abnormal smoke concentrations through gas sensors. The edge computing device can complete data processing locally on the drone and make preliminary judgments, reducing the delay of relying on remote server processing; Communication and warning module: Through 5G, LoRa, and satellite communication technologies, the drone uploads real-time data and processing results to the ground command center. When signs of fire are found, the warning mechanism is automatically triggered to send an alarm to relevant departments through SMS, phone calls, and emails. The command center can receive fire alarm information and respond and dispatch in real time. The ground control and dispatch platform is used to monitor the flight status of the drone and the data it collects in real time. The dispatch platform can display the drone's flight path, sensor data, and fire warning information.

2. According to the edge computing-based forest fire prevention drone patrol warning system of claim 1, it is characterized in that: The drone platform includes a sensor unit, an edge computing device, and a flight control system; Sensor units, mainly including thermal imaging infrared sensors, high-definition visible light cameras, gas sensors, wind speed and direction sensors, which can monitor temperature changes, smoke concentration and other fire-related characteristics in real time; Edge computing devices are equipped on drones to process data from sensors in real time for preliminary identification, analysis and early warning of fires. Edge computing can greatly reduce data transmission delays and improve response speed. The flight control system is used to realize the UAV's autonomous flight, route planning, automatic return and other functions, ensuring that the UAV flies accurately according to the preset patrol path.

3. According to the edge computing-based forest fire prevention drone patrol warning system of claim 1, it is characterized in that: The data acquisition module specifically includes a high-definition camera, an infrared thermal imager, and a meteorological sensor; High-definition cameras can capture high-resolution visible light images to clearly capture forest vegetation conditions, topography, and possible fire sources and smoke anomalies; Infrared thermal imagers can detect the thermal radiation of objects in the forest and effectively identify smoldering fire sources or areas with abnormally high temperatures hidden in the grass and trees. The detection accuracy can reach 0.1°C and can accurately locate the fire source at night or in a dense smoke environment. Meteorological sensors collect meteorological data such as wind speed, wind direction, humidity, and temperature. These data are very important for fire risk assessment and fire spread prediction.

4. The forest fire prevention drone patrol warning system based on edge computing according to claim 1 is characterized in that: The data processing and analysis module specifically includes an edge computing node, a fire prediction unit, and a data fusion and intelligent decision-making unit; Edge computing nodes, after receiving the data transmitted by the drone, the data processing and analysis module immediately processes and analyzes the data. For image and video data, the pre-trained deep learning model is used for target detection to identify possible fire sources and smoke abnormal targets; Fire prediction unit, based on historical data and meteorological data, combined with machine learning or deep learning algorithms, predicts fire risks and provides decision support for fire prevention work; The data fusion and intelligent decision-making unit integrates data from different sensors to improve the accuracy of fire identification. Using big data analysis and pattern recognition technology, the system can accurately predict potential fire risk areas.

5. The forest fire prevention drone patrol warning system based on edge computing according to claim 4 is characterized in that: The edge computing node specifically includes a data processing unit, a data storage unit, and a communication module; The data processing unit is equipped with a high-performance processor to process the images, videos and meteorological data collected by the drone in real time; Data storage unit, used to store a certain amount of historical data and temporary data, including recent patrol images, meteorological data and processing results, so that local data query and analysis can still be performed when the network connection is unstable, and it can also provide data support for subsequent data analysis and model optimization; The communication module is used to realize two-way communication between the UAV and the ground control center, and adopts high-speed and stable wireless communication technology, such as 5G or dedicated microwave communication frequency band, to ensure high-speed data transmission.

6. The forest fire prevention drone patrol warning system based on edge computing according to claim 1 is characterized in that: The communication and warning module specifically includes a wireless communication unit and a warning and response unit; Wireless communication unit, which realizes two-way communication between the UAV and the ground control center. The UAV uploads real-time data and processing results to the ground command center; The early warning and response unit, when the edge computing node detects a fire hazard or determines that a fire has occurred, it immediately generates an early warning message and sends it to the ground control center. After receiving the early warning message, the early warning release system of the ground control center issues an alarm in accordance with the preset early warning method and activates the emergency response mechanism.

7. The forest fire prevention drone patrol warning system based on edge computing according to claim 1 is characterized in that: The ground control and dispatching platform specifically includes a monitoring terminal, a task dispatching and planning unit, and a data management unit; Monitoring terminal: staff can view the flight status, data collection, processing results and warning information of the edge computing node in real time through the monitoring terminal; Task scheduling and planning unit: Based on monitoring data, the command center can schedule tasks for drones, such as adjusting flight paths and increasing patrol frequency; The data management unit manages the massive data generated during the drone patrol process, including data storage, backup, retrieval and analysis. It can mine and analyze historical data, summarize the patterns and characteristics of forest fires, and provide a basis for the formulation of forest fire prevention strategies. The system can also store patrol data and fire warning information in the cloud for subsequent analysis and research.

8. A forest fire prevention drone patrol warning method based on edge computing, characterized in that: The following steps are involved: Step S1. According to the area, topography and key fire prevention areas of the forest area, a patrol mission plan for the drone is formulated at the ground control center; Step S2. After the drone takes off, it flies to the designated location and starts collecting data. The sensors on the drone collect data in real time. The thermal imaging sensor monitors temperature anomalies, the gas sensor monitors smoke concentration, and the wind speed and direction sensor monitors fire propagation conditions. Step S3. The edge computing device on the drone processes the collected data in real time, uses a preset fire recognition algorithm to perform preliminary analysis to determine whether there is a fire hazard. For image and video data, a pre-trained deep learning model is used for target detection to identify possible fire sources and smoke abnormal targets; Step S4. Once a potential fire is discovered, the edge computing node immediately issues an alarm and transmits the alarm information to the ground command center through the wireless communication module. The ground command center responds to the fire risk according to the emergency response. Step S5. After the emergency response is completed, the drone continues its patrol mission and uploads real-time data to the ground control platform for remote monitoring and management; Step S6. After each patrol mission, the ground control center conducts a comprehensive analysis of the data collected by the drone and the processing results of the edge computing node; Step S7. Optimize the deep learning model based on the analysis results. At the same time, update basic information such as map data, vegetation data, and meteorological data of the forest area to provide a more accurate basis for the next patrol mission.

9. The forest fire prevention drone patrol warning method based on edge computing according to claim 8 is characterized in that: Formulating the patrol mission plan of the UAV in step S1 includes determining the take-off point, cruising route, flight altitude, speed and data collection frequency parameters of the UAV, and sending these mission information to the flight control system of the UAV.

10. The forest fire prevention drone patrol warning method based on edge computing according to claim 8 is characterized in that: After the drone takes off in step S2, it flies to the designated location and starts collecting data, including first using a fixed-wing drone to patrol a large area for a long time. Once a suspected fire area is found, a multi-rotor drone is dispatched to conduct a low-altitude detailed patrol of the suspected fire area.

Images