Unmanned aerial vehicle fire extinguishing and rescue system

Abstract

The invention discloses an unmanned aerial vehicle fire extinguishing and rescue system which comprises an emergency command module, a communication navigation emergency command module, an electromagnetic gun window breaking module, a fire extinguishing strong detection module, a rapid rescue module, a self-adaptive building structure analysis module, a self-adaptive deformation structure module and a reconnaissance aircraft module. The invention breaks through the limitation of the traditional fire-fighting operation, solves the problems of height limitation and indoor complex environment maneuvering of the traditional aerial ladder truck, obviously improves the fire rescue efficiency of the super high-rise building, reduces the manual intervention delay, improves the utilization rate of golden rescue time by more than 40%, integrates the functions of window breaking, fire extinguishing and medical rescue, and improves the safety of the super high-rise building. Through millimeter-wave radar life monitoring and precise air-drop, the exposure risk of firemen and trapped persons is reduced, and intelligent and automatic iteration of fire-fighting equipment is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle fire fighting and rescue systems, and in particular to an unmanned aerial vehicle fire fighting and rescue system. Background Art

[0002] In the face of high-rise indoor fires, traditional fire-fighting equipment has obvious limitations, making it difficult to quickly and accurately respond to the fire situation, increasing the risk of fire spread and limiting the rescue efficiency. The following problems exist:

[0003] 1. Limitations of traditional fire-fighting equipment: The aerial ladder truck is limited by height (usually ≤ 50 meters) and cannot quickly reach super high-rise buildings (such as above 35 floors). In indoor fires, the thick smoke and high-temperature environment make it difficult to rescue people.

[0004] 2. Defects of single-function unmanned aerial vehicles: Existing fire-fighting unmanned aerial vehicles are mostly aimed at open scenarios (such as forest fires), lacking the maneuverability and multi-task coordination mechanism for narrow indoor spaces.

[0005] 3. Low information coordination efficiency: The real-time data transmission of the fire scene and the linkage of multiple devices rely on manual decision-making, delaying the golden time for rescue.

[0006] 4. Technical gap: There is no modular unmanned aerial vehicle system integrating functions of breaking windows, fire fighting, and medical rescue, and there is a lack of a full-link automation solution for "reconnaissance - decision - action" for indoor fires.

[0007] Therefore, we propose an unmanned aerial vehicle fire fighting and rescue system to solve the above problems. Summary of the Invention

[0008] The purpose of the present invention is to overcome the deficiencies in the prior art and propose an unmanned aerial vehicle fire fighting and rescue system.

[0009] To achieve the above purpose, the present invention adopts the following technical solutions:

[0010] An unmanned aerial vehicle fire fighting and rescue system includes an emergency command module, a communication and navigation emergency command module, an electromagnetic gun window-breaking module, a fire fighting and strong reconnaissance module, a rapid rescue module, an adaptive building structure analysis module, an adaptive deformation structure module, and a reconnaissance aircraft module.

[0011] Preferably, the emergency command module analyzes fire alarm information through AI algorithms, dynamically dispatching the unmanned aerial vehicle cluster. The emergency command module integrates a fire spread prediction model and generates a rescue path plan in real time. The fire spread prediction model is trained based on historical fire data.

[0012] Preferably, the communication, navigation and emergency command module is based on Beidou positioning and 5G Internet of Things to achieve three-in-one communication among the command center, UAVs and on-site personnel. The communication, navigation and emergency command module adopts an anti-jamming communication protocol to ensure the stability of data transmission in a smoke environment. The communication, navigation and emergency command module is equipped with a multimodal communication link, which includes a main link and a backup link. The main link uses 5GNR-U with a frequency band of 3.5 GHz, and the backup link uses a composite channel of LoRa and laser communication with a bandwidth ≥ 20 MHz.

[0013] Preferably, the electromagnetic gun window-breaking module uses a combination of movable coils and fixed coils, boosts the voltage through a transformer, and uses capacitor energy storage to accelerate the projectile. The projectile uses a high-density alloy warhead to adapt to the payload limit of the UAV. The main control end transmits the three-dimensional fire scene modeling data in real time through 5G to plan the optimal window-breaking path.

[0014] Preferably, the fire extinguishing and strong reconnaissance module includes a reconnaissance unit, an intelligent fire extinguishing agent adaptation system and a fire extinguishing unit. The reconnaissance unit uses an infrared thermal imager and a VOCs gas detector. The reconnaissance unit fuses multi-source data in real time through edge computing to generate a 3D thermal map of the fire scene. The fire extinguishing unit is equipped with a fire extinguishing bomb launching device, which uses a dual-mode control of electromagnetic drive combined with air pressure throttling, and the bombing accuracy is ±0.5 meters. The projectile type of the fire extinguishing bomb launching device is one of dry powder fire extinguishing bombs or cold aerosol bombs. The intelligent fire extinguishing agent adaptation system automatically selects the fire extinguishing bomb type based on the VOCs detection results.

[0015] Preferably, the rapid rescue module is equipped with a medical supplies unit and a vital sign telemetry unit. The medical supplies unit includes a lightweight fireproof backpack, an oxygen mask, an AED defibrillator, and a tourniquet. The delivery mechanism of the rapid rescue module is based on a precise airdrop system using UWB positioning, and the delivery error ≤ 1 meter. The rapid rescue module reserves an interface to connect a portable electrocardiogram monitor to support remote medical consultations. The vital sign telemetry unit integrates a millimeter-wave radar to detect the breathing frequency of the trapped person and predicts the asphyxia risk level through an AI algorithm.

[0016] Preferably, the adaptive building structure analysis module dynamically generates a 3D rescue channel model by inputting the building BIM model, and the adaptive building structure analysis module integrates a fire spread prediction algorithm.

[0017] Preferably, the adaptive deformation structure module includes an arm unit and a rotor unit. The arm unit and the rotor unit use shape memory alloy and can complete the conversion from X-shaped cruise to H-shaped window-breaking structure within 5 seconds. The dynamic adjustment range of the rotor diameter is 30 - 50 cm, adapting to a channel width of 0.6 - 1.2 m.

[0018] Preferably, the reconnaissance aircraft module includes a voice noise reduction unit, an emergency broadcast unit, and a Bluetooth beacon positioning unit. The voice noise reduction unit is equipped with an AI noise reduction chip, which filters out the background noise of the fire scene through voiceprint recognition and directionally picks up the calls for help from trapped people. The emergency broadcast unit accurately broadcasts escape guidance to the trapped area through a parabolic sound wave transmitter. In the Bluetooth beacon positioning unit, the drone throws a low-power Bluetooth beacon, and the mobile phone APP of the trapped person automatically receives the positioning information and uploads it to the command system, with an accuracy within 3 meters.

[0019] The operation process of the present invention is as follows:

[0020] 1. Emergency command and planning: The emergency command module analyzes the fire alarm information with the help of AI algorithms. Based on the fire spread prediction model trained with historical fire data, it predicts the development trend of the fire, and then dynamically schedules the drone cluster to generate a rescue path plan in real time, providing guidance for subsequent rescue operations.

[0021] 2. Communication guarantee: The communication navigation emergency command module uses Beidou positioning and 5G Internet of Things technology to build a three-in-one communication network among the command center, drones, and on-site personnel. It adopts an anti-jamming communication protocol and a multi-modal communication link (the main link is 5GNR-U, and the backup link is a composite channel of LoRa plus laser communication) to ensure the stability of data transmission in harsh environments such as smoke.

[0022] 3. Window-breaking operation: The electromagnetic gun window-breaking module uses a visual recognition algorithm to automatically locate the weak points of the window. Through transformer boosting and capacitor energy storage, it realizes the acceleration of the projectile and can penetrate various types of glass.

[0023] 4. Fire extinguishing and reconnaissance: The reconnaissance unit in the fire extinguishing and strong reconnaissance module uses an infrared thermal imager and a VOCs gas detector to collect fire scene information, and generates a 3D thermal map of the fire scene by real-time fusion of multi-source data through edge computing. The intelligent fire extinguishing agent adaptation system automatically selects the appropriate type of fire extinguishing bomb according to the VOCs detection results. The fire extinguishing bomb launching device of the fire extinguishing unit adopts a dual-mode control of electromagnetic drive combined with air pressure throttling to achieve high-precision bomb dropping.

[0024] 5. Rescue material delivery and life monitoring: The rapid rescue module delivers medical supplies based on a precise airdrop system using UWB positioning, with an error controlled within 1 meter. The medical supply unit is equipped with lightweight fireproof backpacks, oxygen masks, AED defibrillators, tourniquets, etc. The vital sign telemetry unit integrates a millimeter-wave radar to detect the breathing frequency of the trapped person, predicts the asphyxia risk level through AI algorithms, and reserves an interface to support remote medical consultations.

[0025] 6. Rescue passage planning: The adaptive building structure analysis module inputs the building BIM model, combines with the fire spread prediction algorithm, and dynamically generates a 3D rescue passage model to provide feasible rescue paths for drones and rescue personnel.

[0026] 7. Structural adaptive adjustment: The arm units and rotor units of the adaptive deformation structure module are made of shape memory alloy, which can complete the conversion from X-shaped cruise to H-shaped window-breaking structure within 5 seconds, and the rotor diameter can be dynamically adjusted within the range of 30 - 50 cm to adapt to channels of different widths.

[0027] The present invention has the following advantages:

[0028] 1. Solved the problem that traditional aerial ladders are restricted by height and cannot quickly reach super high-rise buildings, can quickly reach the indoor fire scene of high-rise buildings, improved the timeliness of rescue, and reduced the risk of fire spread.

[0029] 2. Designed for narrow indoor spaces, with good maneuverability and multi-task cooperation mechanism, overcame the limitations of single-function drones in indoor rescue, can perform multiple tasks such as window-breaking, fire extinguishing, and rescue in complex indoor environments, and enhanced the indoor combat ability.

[0030] 3. Through advanced communication technology and data processing capabilities, achieved the efficient transmission of real-time fire scene data and the linkage of multiple devices, reduced the delay of manual decision-making, seized the golden rescue time, and improved the information cooperation efficiency.

[0031] 4. Integrated multiple functions such as window-breaking, fire extinguishing, and medical rescue, formed a full-link automation solution for indoor fire detection - decision - action, provided comprehensive and efficient technical support for high-rise indoor fire rescue, and filled the technical gap.

[0032] In summary, the present invention breaks through the limitations of traditional fire fighting operations, solves the problems of the height limitation of traditional aerial ladders and the maneuverability problem in complex indoor environments, significantly improves the efficiency of high-rise building fire rescue, reduces the delay of manual intervention, and the utilization rate of the golden rescue time is increased by more than 40%. It integrates window-breaking, fire extinguishing, and medical rescue functions, reduces the exposure risk of firefighters and trapped people through millimeter-wave radar life monitoring and precise airdrop, and promotes the iteration of fire fighting equipment towards intelligence and automation. Brief Description of the Drawings

[0033] Figure 1 is the system structure diagram of the present invention;

[0034] Figure 2 is the structure diagram of the communication, navigation and emergency command module of the present invention;

[0035] Figure 3 is the structure diagram of the reconnaissance aircraft module of the present invention;

[0036] Figure 4 This is the structural diagram of the fire extinguishing and strong reconnaissance module of the present invention;

[0037] Figure 5 This is the structural diagram of the rapid rescue module of the present invention;

[0038] Figure 6 This is the structural diagram of the adaptive deformation structure module of the present invention. Detailed implementation manners

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0040] Referring to Figure 1-6 , a drone fire extinguishing and rescue system integrates multiple core modules such as emergency command, communication and navigation, window breaking, fire extinguishing reconnaissance, rescue material delivery, rescue channel planning, and structural adaptive adjustment. Each module works together to form an efficient and intelligent rescue system, which can respond in the first time when a fire occurs and minimize casualties and property losses to the greatest extent.

[0041] (1) Emergency command module

[0042] Its core AI algorithm can not only analyze fire alarm information quickly and accurately, but also has certain learning and evolution capabilities. When analyzing fire alarm information, it will comprehensively consider various factors such as the time and location of the fire, building type, and surrounding environment. Based on the fire spread prediction model trained with historical fire data, it will be continuously updated and optimized to adapt to the fire characteristics in different regions and seasons. For example, in the hot and dry summer season, the model will be more conservative in predicting the fire spread speed; while in winter, considering the ventilation situation and distribution of flammable materials in the building, different prediction parameters will be used. By generating real-time rescue path planning, it will not only consider the impact of fire and smoke, but also combine factors such as evacuation channels and distribution of fire fighting facilities in the building to plan the safest and fastest rescue path for the drone cluster.

[0043] (2) Communication navigation emergency command module

[0044] The combination of Beidou positioning and 5G Internet of Things builds an efficient and stable communication bridge among the command center, drones, and on-site personnel. The adoption of the anti-interference communication protocol has been verified through a large number of experiments and actual tests, and can effectively resist the complex electromagnetic interference at the fire scene. The design of the multi-modal communication link is highly forward-looking. The main link uses the 3.5 GHz 5GNR-U frequency band, which has the characteristics of high speed and low latency, and can transmit a large amount of data in real time; the backup link uses a composite channel of LoRa and laser communication. LoRa has the advantage of long-distance communication, and laser communication has better anti-interference ability in complex environments. The combination of the two ensures the continuity and stability of data transmission when the main link fails. The design with a bandwidth ≥ 20 MHz can meet the real-time transmission requirements of various information such as high-definition video and sensor data.

[0045] (III) Electromagnetic gun window-breaking module

[0046] The electromagnetic gun window-breaking module uses a combination of a movable coil and a fixed coil. Through transformer boosting and capacitor energy storage, the projectile is accelerated. The projectile uses a high-density alloy warhead, which is adapted to the payload limit of the drone. The main control end transmits the 3D fire scene modeling data in real time through 5G, plans the optimal window-breaking path, and the control logic is based on a visual recognition algorithm. It can not only automatically locate the weak points of the window, but also dynamically adjust the window-breaking angle and impact force according to factors such as the material and size of the window, ensuring the efficiency and safety of the window-breaking operation.

[0047] (IV) Fire extinguishing and strong reconnaissance module

[0048] The reconnaissance unit uses an infrared thermal imager and a VOCs gas detector, which can collect fire scene information from different angles. The infrared thermal imager can monitor the temperature distribution of the fire scene in real time, discover potential fire sources and the direction of fire spread; the VOCs gas detector can detect volatile organic compounds in the air and judge the type of fire and the burning substances. Through edge computing to fuse multi-source data in real time, the generated 3D thermal map of the fire scene is more accurate and intuitive, providing strong support for fire extinguishing decision-making. The intelligent fire extinguishing agent adaptation system automatically selects the type of fire extinguishing bomb based on the VOCs detection results, and can select the most suitable dry powder fire extinguishing bomb or cold aerosol bomb according to different types of fires, such as oil fires and electrical fires, to improve the fire extinguishing efficiency. The fire extinguishing bomb launching device of the fire extinguishing unit uses a dual-mode control of electromagnetic drive combined with air pressure throttling. This innovative control method can achieve high-precision bomb throwing, and the bomb throwing accuracy reaches ±0.5 meters, ensuring that the fire extinguishing bomb can accurately hit the fire source.

[0049] (V) Quick rescue module

[0050] Medical supplies such as lightweight fireproof backpacks, oxygen masks, AED defibrillators, and tourniquets inside the medical supply unit are all carefully selected and configured to meet the basic needs of emergency rescue at the fire scene. The design of the lightweight fireproof backpack facilitates the carrying and dropping by drones; the oxygen mask can provide timely oxygen support for trapped people; the AED defibrillator can perform first aid on patients with cardiac arrest in case of emergency; and the tourniquet can be used to control bleeding from wounds. The precise airdrop system based on UWB positioning can achieve high-precision dropping in complex building environments, with a dropping error ≤ 1 meter. The vital sign telemetry unit integrates a millimeter-wave radar to detect the breathing frequency of trapped people, predicts the asphyxia risk level through an AI algorithm, can timely detect the life danger of trapped people, and provides accurate data support for remote medical consultations. The reserved interface is connected to a portable electrocardiogram monitor to support remote medical consultations, enabling professional doctors to understand the vital signs of trapped people in real time at the command center and guide on-site rescue personnel to perform first aid.

[0051] (VI) Adaptive building structure analysis module

[0052] By inputting the building BIM model, this module can obtain detailed structural information of the building, including floor layout, wall thickness, door and window positions, etc. Combining with the fire spread prediction algorithm, the dynamically generated 3D rescue channel model is more accurate and practical. When generating the rescue channel model, factors such as the development trend of the fire and the diffusion direction of the smoke will be considered to plan the safest and most feasible rescue path for drones and rescue personnel. For example, in the initial stage of the fire, channels farther from the fire source may be preferentially selected; while when the fire is larger, rescue may be considered through special channels such as ventilation ducts.

[0053] (VII) Adaptive deformation structure module

[0054] The arm unit and the rotor unit of the drone adopt shape memory alloy, which has unique physical properties and can change its shape under different temperature and stress conditions. It can complete the conversion from the X-shaped cruise to the H-shaped window-breaking structure within 5 seconds, enabling the drone to maintain good stability and maneuverability during the cruise and quickly convert to a structure suitable for window-breaking operations when reaching the window-breaking position. The dynamic adjustment range of the rotor diameter is 30 - 50 cm, which can adapt to a channel width of 0.6 - 1.2 m, allowing the drone to freely shuttle through building channels of different sizes and improving the flexibility and adaptability of rescue.

[0055] (VIII) Reconnaissance aircraft module

[0056] The reconnaissance aircraft module includes a voice noise reduction unit, an emergency broadcast unit, and a Bluetooth beacon positioning unit. The voice noise reduction unit is equipped with an AI noise reduction chip, which filters out the background noise of the fire scene through voiceprint recognition and directionally picks up the calls for help from trapped people. The emergency broadcast unit accurately broadcasts escape guidance to the trapped area through a parabolic sound wave transmitter. In the Bluetooth beacon positioning unit, the drone throws a low-power Bluetooth beacon, and the mobile phone APP of the trapped person automatically receives the positioning information and uploads it to the command system, with an accuracy of within 3 meters.

[0057] The operation process of the present invention:

[0058] (I) Emergency command and planning

[0059] At the first moment of a fire, the emergency command module will be immediately activated. The AI algorithm quickly collects fire alarm information, including the content of fire alarm calls, surrounding surveillance videos, etc. At the same time, it calls the fire spread prediction model trained based on historical fire data to predict the development trend of the fire. During the prediction process, real-time meteorological data, such as wind speed, wind direction, humidity, etc., and the structural characteristics of the building, such as ventilation conditions, distribution of flammable materials, etc., will be combined to make a more accurate prediction. According to the prediction results, the drone cluster is dynamically dispatched, and specific tasks and paths are assigned to each drone. When generating the rescue path plan, factors such as the flight performance and endurance of the drone will be considered to ensure the feasibility and safety of the path.

[0060] (II) Communication guarantee

[0061] The communication navigation emergency command module quickly constructs a three-in-one communication network among the command center, drones, and on-site personnel while the emergency command module is activated. The Beidou positioning system provides accurate position information for the drones to ensure that they can accurately reach the designated locations. The 5G Internet of Things enables high-speed data transmission, allowing the command center to real-time monitor the situation of drones and on-site personnel. The anti-interference communication protocol automatically identifies and filters out electromagnetic interference signals at the fire scene to ensure the stability of data transmission. The multi-modal communication link automatically switches according to the actual situation. When the main link 5GNR-U is interfered with or the signal is interrupted, the backup link LoRa plus laser communication composite channel will be immediately activated to ensure the continuity of communication.

[0062] (III) Window-breaking operation

[0063] The electromagnetic gun window-breaking module starts working after the drone reaches the designated position. The visual recognition algorithm quickly scans and analyzes the window to locate the weak points of the window. At the same time, according to the material and size of the window, the window-breaking angle and impact force are automatically adjusted. The glass stress detection unit monitors the expansion of glass cracks in real time. When the cracks expand to a certain extent, the power system dynamically adjusts the window-breaking impact force to ensure that the window can be broken smoothly. During the window-breaking process, the attitude of the drone remains stable to improve the accuracy and safety of window-breaking.

[0064] (IV) Fire extinguishing and reconnaissance

[0065] The fire extinguishing and strong reconnaissance module starts working immediately after the window-breaking is completed. The infrared thermal imager and VOCs gas detector of the reconnaissance unit are started simultaneously to collect information on the temperature distribution and gas composition in the fire scene. Through edge computing, multi-source data are fused in real time to generate a 3D thermal map of the fire scene. The intelligent fire extinguishing agent adaptation system automatically selects the appropriate type of fire extinguishing bomb according to the VOCs detection results. The fire extinguishing bomb launching device of the fire extinguishing unit adjusts the launching angle and force according to the 3D thermal map of the fire scene and the target position, and adopts the dual-mode control of electromagnetic drive combined with air pressure throttling to achieve high-precision bomb dropping. During the bomb dropping process, the bomb dropping parameters are continuously adjusted to ensure that the fire extinguishing bomb can accurately hit the fire source.

[0066] (V) Rescue material delivery and life monitoring

[0067] The rapid rescue module starts to carry out rescue material delivery and life monitoring while the fire extinguishing work is in progress. The precise airdrop system based on UWB positioning accurately calculates the delivery parameters according to the position information of the trapped people to ensure that medical supplies can be accurately delivered near the trapped people. The high-precision requirement of the delivery error ≤ 1 meter can ensure that the trapped people can obtain rescue supplies in time. The millimeter-wave radar of the vital sign telemetry unit monitors the trapped people in real time to detect their breathing frequency. Through the AI algorithm, the breathing frequency data are analyzed to predict the asphyxia risk level of the trapped people. If a high asphyxia risk is found, an alarm message will be immediately sent to the command center through the communication network, and at the same time, data support will be provided for remote medical consultation.

[0068] (VI) Rescue channel planning

[0069] The adaptive building structure analysis module works continuously throughout the rescue process. By inputting the building BIM model and combining the real-time fire spread situation and the information collected by the reconnaissance unit, a 3D rescue channel model is dynamically generated. When generating the model, the fire and smoke situations on different floors and in different areas are taken into account to plan multiple feasible rescue paths for drones and rescue personnel. At the same time, the rescue channel model is adjusted in real time according to the progress of the rescue work to ensure the safety and effectiveness of the rescue path.

[0070] (7) Structural adaptive adjustment

[0071] The adaptive deformation structure module makes structural adjustments in a timely manner according to the needs of rescue missions. During the cruise of the UAV, the arm unit and the rotor unit maintain an X-shaped structure to improve the flight stability and maneuverability. When the UAV reaches the window-breaking position or needs to pass through a narrow passage, the arm unit and the rotor unit will complete the conversion from the X-shaped cruise structure to the H-shaped window-breaking structure within 5 seconds. At the same time, the rotor diameter will be dynamically adjusted according to the width of the passage to ensure that the UAV can smoothly pass through passages of different sizes.

[0072] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, should be covered by the protection scope of the present invention.

Claims

1. A UAV firefighting and rescue system, including an emergency command module, a communication and navigation emergency command module, an electromagnetic gun window breaking module, a firefighting and strong detection module, a rapid rescue module, an adaptive building structure analysis module, an adaptive deformation structure module and a reconnaissance aircraft module.

2. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The emergency command module analyzes fire alarm information through AI algorithms and dynamically dispatches drone clusters. The emergency command module integrates a fire spread prediction model to generate rescue path planning in real time. The fire spread prediction model is trained based on historical fire data.

3. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The communication and navigation emergency command module is based on Beidou positioning and 5G Internet of Things to achieve three-in-one communication between the command center, drone and on-site personnel. The communication and navigation emergency command module adopts an anti-interference communication protocol to ensure the stability of data transmission in a smoky environment. A multimodal communication link is provided in the communication and navigation emergency command module. The multimodal communication link includes a main link and a backup link. The main link adopts 5GNR-U with a frequency band of 3.5GHz, and the backup link adopts a LoRa plus laser communication composite channel with a bandwidth ≥20MHz.

4. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The electromagnetic gun window breaking module adopts a combination of movable coils and fixed coils, boosts the voltage through a transformer, and uses capacitor energy storage to accelerate the projectile. The projectile adopts a high-density alloy warhead to adapt to the load limit of the drone. The main control end transmits the three-dimensional modeling data of the fire scene in real time through 5G to plan the optimal window breaking path.

5. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The fire extinguishing and detection module includes a reconnaissance unit, an intelligent fire extinguishing agent adaptation system and a fire extinguishing unit. The reconnaissance unit adopts an infrared thermal imager and a VOCs gas detector. The reconnaissance unit fuses multi-source data in real time through edge computing to generate a 3D thermal map of the fire scene. A fire extinguishing bomb launching device is provided in the fire extinguishing unit. The fire extinguishing bomb launching device adopts electromagnetic drive combined with air pressure throttling dual-mode control, with a bomb throwing accuracy of ±0.5 meters. The bomb body type of the fire extinguishing bomb launching device is one of dry powder fire extinguishing bombs or cold aerosol bombs. The intelligent fire extinguishing agent adaptation system automatically selects the type of fire extinguishing bomb according to the VOCs detection results.

6. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The rapid rescue module is equipped with a medical supplies unit and a vital signs telemetry unit. The medical supplies unit includes a lightweight fireproof backpack, an oxygen mask, an AED defibrillator, and a tourniquet. The delivery mechanism of the rapid rescue module is based on a precise airdrop system based on UWB positioning, and the delivery error is ≤1 meter. The rapid rescue module has a reserved interface for connecting a portable ECG monitor to support remote medical consultation. The vital signs telemetry unit integrates a millimeter-wave radar to detect the respiratory rate of the trapped person and predicts the suffocation risk level through an AI algorithm.

7. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The adaptive building structure analysis module dynamically generates a 3D rescue channel model by inputting a building BIM model, and the adaptive building structure analysis module integrates a fire spread prediction algorithm.

8. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The adaptive deformation structure module includes an arm unit and a rotor unit. The arm unit and the rotor unit are made of shape memory alloy and can complete the conversion from X-type cruise to H-type broken window structure within 5 seconds. The dynamic adjustment range of the rotor diameter is 30-50cm, which can adapt to the channel width of 0.6-1.2m.

9. The unmanned aerial vehicle firefighting and rescue system according to claim 1, characterized in that: The reconnaissance aircraft module includes a voice noise reduction unit, an emergency broadcast unit and a Bluetooth beacon positioning unit. The voice noise reduction unit is equipped with an AI noise reduction chip, which filters the background noise of the fire scene through voiceprint recognition and directionally picks up the cries for help of trapped people. The emergency broadcast unit uses a parabolic sound wave transmitter to accurately broadcast escape instructions to the trapped area. The drone in the Bluetooth beacon positioning unit throws a low-power Bluetooth beacon, and the trapped person’s mobile phone APP automatically receives the positioning information and uploads it to the command system with an accuracy of within 3 meters.

Images