A method for monitoring the living environment of a drone and a system therefor

Abstract

This invention discloses a method and system for monitoring human settlements using unmanned aerial vehicles (UAVs), relating to the field of human settlements monitoring. The method includes setting up a monitoring module mounted on a UAV, establishing a real-time communication link between the module and the flight control system, enabling the UAV to fly over human settlements, and during flight, the monitoring module collects parameters of the human settlements. Based on the flight control data, the collected parameters of the monitoring module are dynamically corrected to generate a calibrated data set. The calibrated data is then input into a scenario-based analytical model. This invention utilizes a real-time correction mechanism with deep coupling between flight control and sensing to dynamically correlate UAV attitude parameters with monitoring data, thereby reducing the data error rate and eliminating the need for post-correction. This is particularly beneficial in complex terrains such as mountainous blind areas and urban streets, improving the accuracy of pollutant concentration monitoring and helping to solve the problem of data distortion caused by dynamic interference.

Description

Technical Field

[0001] This invention relates to the field of human settlement environment monitoring, and in particular to a method and system for human settlement environment monitoring using unmanned aerial vehicles (UAVs). Background Technology

[0002] Drones are aircraft that do not require direct human pilot control and fly through remote control, autonomous programs, or artificial intelligence systems. Drones have a wide range of applications, including monitoring of human settlements.

[0003] When using drones to monitor human settlements, the flight attitude of existing drones is usually physically isolated from sensor data collection. Environmental interference can only be corrected through post-processing, which cannot respond in real time to dynamic factors such as low airflow disturbances and fuselage vibrations. This leads to high error rates for core parameters such as pollutant concentration and soil moisture. Therefore, it is necessary to use a drone-based human settlement monitoring method that can dynamically correct interference in real time and integrate scenario-based data landing. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for monitoring human settlements using unmanned aerial vehicles (UAVs) to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring human settlement environment using an unmanned aerial vehicle (UAV), comprising: Step 1: Set up the monitoring module and mount it on the UAV. Then establish a real-time communication link between the module and the flight control system. The monitoring module integrates gas sensing, visual imaging and attitude perception. Step 2: The drone flies over the human settlement environment, and during the flight, the monitoring module collects parameters of the human settlement environment. Then, based on the flight control data, the collected parameters of the monitoring module are dynamically corrected to generate a calibrated data set. Step 3: Input the calibration data from Step 2 into the scenario-based analysis model, then output the quantitative indicators of potential hazards in the living environment and present the visualization results; Step 4: Then, based on the analysis results in Step 3, trigger the flight control system to adjust the flight path or the ground terminal to issue an early warning.

[0006] Preferably, the monitoring module in step one is compatible with the MAVLink communication protocol and synchronizes data with the flight control system via a TTL interface at the millisecond level. The attitude perception and acquisition of UAV parameters in step one specifically includes roll angle, vibration frequency, and altitude change rate parameters.

[0007] Preferably, the calibration data in step two specifically includes: S1: The flight control system outputs flight status data in real time, including instantaneous speed, attitude angle, and power system load; S2: Calculate the dynamic interference correction coefficient The formula for calculating the dynamic interference correction coefficient is as follows: ; in, This refers to the sensor type weight. This refers to the altitude correction factor; S3: Combine the original monitoring data with the correction coefficient Perform convolution operations to obtain calibration data.

[0008] Preferably, in step S2 The value is set according to the monitoring parameter type: If it is a gas concentration parameter, then The value is 0.8; If it is a visual grayscale value parameter, then The value is 0.5; If it is a soil moisture parameter, then The value is 0.6.

[0009] Preferably, in step S3, a sliding window algorithm is used to smooth the convolution result, and the window width is dynamically adjusted according to the flight speed. When the flight speed is ≥10m / s, the window width is set to 50ms; When 5m / s ≤ flight speed < 10m / s, the window width is set to 35ms; When the flight speed is less than 5 m / s, the window width is set to 20 ms, and the step size of the sliding window is fixed at 10 ms.

[0010] Preferably, the scenario-based analysis model in step three pre-stores three types of human settlement environment scenario libraries, namely garbage accumulation, water pollution and pipeline leakage, and each scenario contains at least five sets of related parameter threshold tables.

[0011] Preferably, step three specifically includes: SS1: Perform feature extraction on calibration data, remove environmental noise interference through filtering algorithms, and identify core parameter types, including pollutant concentration, turbidity, characteristic gases, and visual texture features; SS2: Based on the core parameter type matching corresponding human settlement environment scenario library, a multi-dimensional evaluation matrix is ​​constructed using the analytic hierarchy process (AHP). The hazard matching degree is calculated by combining scenario weight coefficients. The scenario weight coefficients are dynamically configured according to the regional population density, environmental functional zoning, and distribution of sensitive points. SS3: Determine the risk level based on the threshold range of the hazard matching degree. The matching degree threshold specifically includes: if the hazard matching degree threshold is ≥85%, it is determined to be high risk; if the hazard matching degree threshold is between 60% and 84%, it is determined to be medium risk; if the hazard matching degree threshold is <60%, it is determined to be low risk. Then, a quantitative indicator system containing hazard level, impact range, rectification priority, and disposal suggestions is generated, and a three-dimensional visualization heat map and geographic coordinate mark file are generated simultaneously.

[0012] Preferably, step four further includes a dual threshold judgment mechanism, specifically including: When the calibrated pollutant concentration exceeds the first-level threshold, the analytical model is triggered to upgrade to emergency mode, and then the parameter acquisition frequency is increased to 50Hz. When the hazard level reaches level three or above, the flight control system will automatically plan a secondary inspection route and perform a low-altitude scan of the hazard area at a height of 10m.

[0013] Preferably, the early warning feedback in step four includes pushing a data packet containing latitude and longitude, quantitative indicators, and visualization results to the ground terminal. The data packet is transmitted using AES-256 encryption and supports import. In step one, after establishing the communication link between the monitoring module and the flight control system, the GPS positioning module and terrain preprocessing data carried by the UAV are used to determine whether the monitoring area meets the collaborative triggering conditions. The triggering conditions are that when the monitoring area is ≥5km², or the terrain preprocessing data shows that there are monitoring blind spots with building obstruction and terrain drop ≥30m, the collaborative monitoring mode of multiple UAVs is activated. In the dynamic correction process in step two, each UAV synchronizes flight control status data and dynamic interference correction coefficient in real time through distributed communication nodes. Then, the cross-machine collected data is calibrated a second time.

[0014] A drone-based human settlement environment monitoring system includes: The monitoring module is used to set up a monitoring module and mount it on the UAV, and then establish a real-time communication link between the module and the flight control system. The monitoring module integrates gas sensing, visual imaging and attitude perception. The flight control unit is used to enable the UAV to fly over the human settlement environment. During the flight, the monitoring module collects parameters of the human settlement environment and then dynamically corrects the collected parameters of the monitoring module based on the flight control data to generate a calibrated data set. The scenario analysis module is used to input calibration data into the scenario-based analysis model, and then output quantitative indicators of potential hazards in the human living environment and present the visualization results. The early warning feedback module is used to trigger the flight control system to adjust the flight path or the ground terminal to provide early warning feedback based on the analysis results.

[0015] The technical effects and advantages of this invention are as follows: This invention uses a real-time correction mechanism that deeply couples flight control and sensing to dynamically correlate UAV attitude parameters with monitoring data, thereby reducing the error rate of the data and eliminating the need for post-correction. This is especially beneficial in complex terrains such as mountainous blind areas and urban streets, as it helps to improve the accuracy of pollutant concentration monitoring and thus helps to solve the problem of data distortion caused by dynamic interference. This invention transforms raw data into intuitive indicators such as hazard level and rectification priority by using a scenario-based analytical model. Combined with the output of a 3D heat map, it can be directly used for decision-making on the improvement of the living environment, thereby reducing the cost of human interpretation and shortening the response time for hazard handling. It also helps to break down the professional barriers to data implementation and facilitates operation. This invention provides reliable input for scene analysis by utilizing corrected and accurate data. This helps to identify high-priority hidden dangers, which in turn drive the flight control to automatically adjust the flight path. This, in turn, helps to form a closed loop of data collection, correction, analysis, and re-collection, making it easier to upgrade UAVs from data collection tools to intelligent decision-making terminals. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the process for monitoring the human settlement environment using an unmanned aerial vehicle (UAV) according to the present invention. Figure 2 This is a schematic diagram of the specific process of step three of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] This invention provides, for example Figures 1-2 The method for monitoring human settlements using unmanned aerial vehicles (UAVs) includes the following specific steps: Step 1: Set up the monitoring module and mount it on the drone. Then, establish a real-time communication link between the module and the flight control system. The monitoring module integrates gas sensing, visual imaging, and attitude perception. It uses a quick-release bracket to connect to the drone fuselage, with built-in shock-absorbing pads. The installation position is directly below the drone's center of gravity to prevent module displacement during flight attitude adjustments. The air inlets of the gas sensing units (such as the electrochemical CO sensor and infrared ammonia sensor) face the flight direction to reduce the impact of airflow turbulence on sampling efficiency. The lenses of the visual imaging units (high-definition camera and infrared thermal imaging module) point vertically downwards, ensuring the ground monitoring range covers 1.5 times the width of the drone's flight path. Electrical interface connection: The module draws power from the drone's power system via a DC-DC step-down module, with a supply current ≥2A to meet the power consumption requirements of multiple sensors operating simultaneously. The communication interface uses a TTL serial port to interface with the flight control system's serial port. Ensure correct wiring during connection. Negative polarity reverse protection prevents voltage fluctuations from damaging the module; Monitoring module initialization and parameter configuration operations, power-on self-test process: After the UAV starts, the monitoring module automatically enters self-test mode, sequentially checking the working status of each unit: the gas sensing unit performs zero-point calibration, the visual imaging unit outputs test images, and the attitude perception unit (three-axis gyroscope and accelerometer) outputs initial attitude data (tilt angle and vibration frequency default values ​​are 0); after passing the self-test, the monitoring module sends a ready signal to the flight control system via the MAVLink protocol. After the flight control system confirms the signal, the module enters standby mode, waiting for acquisition commands; Parameter preset configuration: Based on the monitoring task type (such as garbage accumulation monitoring and water pollution monitoring), the sensor acquisition parameters are preset through the ground terminal: the gas sensing unit sampling frequency is 10Hz by default, the visual imaging unit shooting interval is 2 seconds / frame, and the attitude perception unit data output frequency is 50Hz; Preset sensor type weights. and altitude correction factor The data is stored in the module's built-in chip; dynamic acquisition and collaborative operation during flight, real-time data acquisition: After the UAV takes off, the module starts data acquisition according to the flight control system's flight commands: during cruise, data is acquired at a preset frequency, and the acquisition frequency is increased to 20Hz when hovering; the gas sensing unit acquires raw data on pollutant concentration, the visual imaging unit simultaneously captures ground images and extracts grayscale values ​​and texture features, and the attitude sensing unit acquires the UAV's roll angle, vibration frequency, and altitude change rate in real time; collaborative communication with the flight control system: every 10ms, the monitoring module sends the acquired attitude data to the flight control system via the MAVLink protocol. After receiving the data, the flight control system combines it with its own flight status data (instantaneous speed, power load) and feeds back dynamic adjustment commands (such as adjusting the gas sensor sampling gain and correcting the visual imaging exposure parameters) through the same link; during the dynamic correction process, the monitoring module receives the correction coefficient sent by the flight control system. The raw data is processed through the built-in MCU. The convolution operation generates calibration data, which is temporarily stored in the cache for later use by the scenario-based parsing model. If a sensor malfunction occurs during the acquisition process (such as the gas sensing unit data exceeding the measurement range or the visual imaging unit not outputting images), the monitoring module immediately sends an abnormal signal to the flight control system and automatically switches to the backup sensor (such as activating the backup sensor when the main CO sensor fails). If the communication link is interrupted (no feedback is received from the flight control system for 3 consecutive times), the monitoring module enters an independent working mode and continues to acquire data according to preset parameters. After the link is restored, the cached data is automatically synchronized to ensure data integrity. Step 2: The drone flies over the human settlement environment, and during the flight, the monitoring module collects parameters of the human settlement environment. Then, based on the flight control data, the collected parameters of the monitoring module are dynamically corrected to generate a calibrated data set. Step 3: Input the calibration data from Step 2 into the scenario-based analysis model, then output the quantitative indicators of potential hazards in the living environment and present the visualization results; Step 4: Then, based on the analysis results in Step 3, trigger the flight control system to adjust the flight path or the ground terminal to issue an early warning.

[0019] Furthermore, the monitoring module in step one is compatible with the MAVLink communication protocol. MAVLink is a lightweight, low-bandwidth communication protocol designed for micro-aircraft. It uses a binary message format, with a single message length of only 8-255 bytes, a transmission latency of ≤10ms, and supports half-duplex communication. It can transmit at multiple physical layers such as serial ports (e.g., TTL interfaces), CAN buses, and UDP, adapting to the complex electromagnetic environment of UAVs flying at low altitudes. The monitoring module and the flight control system establish a MAVLink communication link via a TTL interface. The protocol message frame contains the following core fields: Frame header: identifies the start of the message; Payload length: stores attitude sensing data (tilt angle, vibration frequency, and altitude change rate); System ID / Component ID. D: Identifies the UAV platform (System ID=1) and monitoring module (Component ID=20) respectively; Message ID: Defines 3 types of messages (ID=101: Tilt angle data, ID=102: Vibration frequency data, ID=103: Altitude change rate data); Checksum: Ensures data transmission integrity; Synchronization mechanism: The flight control system sends an attitude data message every 10ms. After receiving the message, the monitoring module confirms it through the MAVLink ACK feedback mechanism, achieving millisecond-level data synchronization and providing real-time data support for the dynamic correction in step two; and it synchronizes the data with the flight control system through the TTL interface at the millisecond level. The attitude perception and acquisition of UAV parameters in step one specifically includes tilt angle, vibration frequency, and altitude change rate parameters.

[0020] Furthermore, the calibration data in step two specifically includes: S1: The flight control system outputs flight status data in real time, including instantaneous speed, attitude angle, and power system load; S2: Calculate the dynamic interference correction coefficient The formula for calculating the dynamic interference correction coefficient is as follows: ; in, This refers to the sensor type weight. This refers to the altitude correction factor; S3: Combine the original monitoring data with the correction coefficient Perform convolution operations to obtain calibration data.

[0021] Specifically, in step S2 The value is set according to the monitoring parameter type: If it is a gas concentration parameter, then The value is 0.8; If it is a visual grayscale value parameter, then The value is 0.5; If it is a soil moisture parameter, then The value is 0.6.

[0022] Furthermore, in step S3, a sliding window algorithm is used to smooth the convolution result, and the window width is dynamically adjusted according to the flight speed. When the flight speed is ≥10m / s, the window width is set to 50ms; When 5m / s ≤ flight speed < 10m / s, the window width is set to 35ms; When the flight speed is <5m / s, the window width is set to 20ms, and the step size of the sliding window is fixed at 10ms to ensure the continuity and real-time performance of data smoothing. The sliding window algorithm is a time-series data processing method for data smoothing. It calculates a rolling average of continuously collected calibration data by setting a fixed-length window, eliminating instantaneous noise interference while preserving the dynamic trend of the data. Combined with the implementation logic of step S23, the operation flow of the sliding window algorithm is as follows: Initialize window parameters. Based on the set speed threshold, preset three sets of window parameters: when the flight speed is ≥10m / s, the window width = 50ms (i.e., containing 5 sampling points, sampling frequency = 100Hz); when 5m / s ≤ flight speed 0m / s, the window width = 35ms (containing 3.5 sampling points, using linear interpolation for completion); when the flight speed... When the speed is <5m / s, the window width is 20ms (including 2 sampling points); the window step size is uniformly set to 10ms (i.e., the window slides once every 10ms to ensure data continuity); data is fed into the window, and the calibration data after convolution operation is sequentially fed into the sliding window in chronological order, with the data in the window being updated in real time (when new data enters, the oldest data is removed); smoothing calculation is performed, and the output value is calculated using a weighted average method for all data in the window: let the i-th data in the window be x_i (i=1,2,...,n, where n is the number of data in the window), and the corresponding weight be w_i (w_i=1 / n, i.e., equal weighted average, which simplifies the calculation and ensures real-time performance), then the smoothed data y=Σ(x_i×w_i); output results are generated after each window slide, forming a continuous and stable calibration dataset for scenario-based analysis in the subsequent step three.

[0023] Furthermore, in step three, the scenario-based analysis model pre-stores three types of human settlement environment scenario libraries: garbage accumulation, water pollution, and pipeline leakage. Each scenario contains at least five sets of related parameter threshold tables.

[0024] Specifically, step three includes: SS1: Perform feature extraction on calibration data, remove environmental noise interference through filtering algorithms, and identify core parameter types, including pollutant concentration, turbidity, characteristic gases, and visual texture features; SS2: Based on the core parameter type matching corresponding human settlement environment scenario library, a multi-dimensional evaluation matrix is ​​constructed using the analytic hierarchy process. The hazard matching degree is calculated by combining scenario weight coefficients. The scenario weight coefficients are dynamically configured according to the regional population density, environmental functional zoning and distribution of sensitive points. SS3: The risk level is determined based on the threshold range of the hazard matching degree. The matching degree threshold is as follows: if the hazard matching degree threshold is ≥85%, it is judged as high risk; if the hazard matching degree threshold is between 60% and 84%, it is judged as medium risk; and if the hazard matching degree threshold is <60%, it is judged as low risk. Then, a quantitative indicator system containing hazard level, impact range, rectification priority and disposal suggestions is generated, and a three-dimensional visualization heat map and geographic coordinate mark file are generated simultaneously.

[0025] Furthermore, step four also includes a dual threshold determination mechanism, specifically including: When the calibrated pollutant concentration exceeds the first-level threshold, the analytical model is triggered to upgrade to emergency mode, and then the parameter acquisition frequency is increased to 50Hz. When the hazard level reaches level three or above, the flight control system will automatically plan a secondary inspection route and perform a low-altitude scan of the hazard area at a height of 10m.

[0026] Specifically, the early warning feedback in step four includes pushing data packets containing latitude and longitude, quantitative indicators, and visualization results to the ground terminal. The data packets are transmitted using AES-256 encryption and can be imported. After establishing the communication link between the monitoring module and the flight control system in step one, the GPS positioning module and terrain preprocessing data on the UAV are used to determine whether the monitoring area meets the collaborative triggering conditions. The triggering conditions are that when the monitoring area is ≥5km², or when the terrain preprocessing data shows that there are monitoring blind spots with building obstruction and terrain drop ≥30m, the collaborative monitoring mode of multiple UAVs is activated. In the dynamic correction process in step two, each UAV synchronizes flight control status data and dynamic interference correction coefficients in real time through distributed communication nodes. Then, the cross-machine data collection is calibrated a second time to ensure the consistency of data from multiple machines within the same monitoring area, so that the cross-machine data deviation rate is controlled within ±2%.

[0027] A drone-based human settlement environment monitoring system includes: The monitoring module is used to set up a monitoring module and mount it on the UAV, and then establish a real-time communication link between the module and the flight control system. The monitoring module integrates gas sensing, visual imaging and attitude perception. The flight control unit is used to enable the UAV to fly over the human settlement environment. During the flight, the monitoring module collects parameters of the human settlement environment and then dynamically corrects the collected parameters of the monitoring module based on the flight control data to generate a calibrated data set. The scenario analysis module is used to input calibration data into the scenario-based analysis model, and then output quantitative indicators of potential hazards in the human living environment and present the visualization results. The early warning feedback module is used to trigger the flight control system to adjust the flight path or the ground terminal to provide early warning feedback based on the analysis results.

[0028] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for monitoring human settlements using unmanned aerial vehicles (UAVs), characterized in that: The specific steps include the following: Step 1: Set up the monitoring module and mount it on the UAV. Then establish a real-time communication link between the module and the flight control system. The monitoring module integrates gas sensing, visual imaging and attitude perception. Step 2: The drone flies over the human settlement environment, and during the flight, the monitoring module collects parameters of the human settlement environment. Then, based on the flight control data, the collected parameters of the monitoring module are dynamically corrected to generate a calibrated data set. Step 3: Input the calibration data from Step 2 into the scenario-based analysis model, then output the quantitative indicators of potential hazards in the living environment and present the visualization results; Step 4: Then, based on the analysis results in Step 3, trigger the flight control system to adjust the flight path or the ground terminal to issue an early warning.

2. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The monitoring module in step one is compatible with the MAVLink communication protocol and synchronizes data with the flight control system via a TTL interface at the millisecond level. The attitude perception and acquisition of UAV parameters in step one specifically includes roll angle, vibration frequency, and altitude change rate parameters.

3. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The calibration data in step two specifically includes: S1: The flight control system outputs flight status data in real time, including instantaneous speed, attitude angle, and power system load; S2: Calculate the dynamic interference correction coefficient The formula for calculating the dynamic interference correction coefficient is as follows: ； in, This refers to the sensor type weight. This refers to the altitude correction factor; S3: Combine the original monitoring data with the correction coefficient Perform convolution operations to obtain calibration data.

4. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: In step S2 The value is set according to the monitoring parameter type: If it is a gas concentration parameter, then The value is 0.8; If it is a visual grayscale value parameter, then The value is 0.5; If it is a soil moisture parameter, then The value is 0.

6.

5. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: In step S3, a sliding window algorithm is used to smooth the convolution result, and the window width is dynamically adjusted according to the flight speed. When the flight speed is ≥10m / s, the window width is set to 50ms; When 5m / s ≤ flight speed < 10m / s, the window width is set to 35ms; When the flight speed is less than 5 m / s, the window width is set to 20 ms, and the step size of the sliding window is fixed at 10 ms.

6. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The scenario-based analysis model in step three pre-stores three types of human settlement environment scenario libraries, namely garbage accumulation, water pollution and pipeline leakage, and each scenario contains at least five sets of related parameter threshold tables.

7. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: Step three specifically includes: SS1: Perform feature extraction on calibration data, remove environmental noise interference through filtering algorithms, and identify core parameter types, including pollutant concentration, turbidity, characteristic gases, and visual texture features; SS2: Based on the core parameter type matching corresponding human settlement environment scenario library, a multi-dimensional evaluation matrix is ​​constructed using the analytic hierarchy process (AHP). The hazard matching degree is calculated by combining scenario weight coefficients. The scenario weight coefficients are dynamically configured according to the regional population density, environmental functional zoning, and distribution of sensitive points. SS3: Determine the risk level based on the threshold range of the hazard matching degree. The matching degree threshold specifically includes: if the hazard matching degree threshold is ≥85%, it is determined to be high risk; if the hazard matching degree threshold is between 60% and 84%, it is determined to be medium risk; if the hazard matching degree threshold is <60%, it is determined to be low risk. Then, a quantitative indicator system containing hazard level, impact range, rectification priority, and disposal suggestions is generated, and a three-dimensional visualization heat map and geographic coordinate mark file are generated simultaneously.

8. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: Step four also includes a dual threshold judgment mechanism, specifically including: When the calibrated pollutant concentration exceeds the first-level threshold, the analytical model is triggered to upgrade to emergency mode, and then the parameter acquisition frequency is increased to 50Hz. When the hazard level reaches level three or above, the flight control system will automatically plan a secondary inspection route and perform a low-altitude scan of the hazard area at a height of 10m.

9. The method for monitoring human settlement environment using an unmanned aerial vehicle (UAV) according to claim 2, characterized in that: The early warning feedback in step four includes pushing a data packet containing latitude and longitude, quantitative indicators, and visualization results to the ground terminal. This data packet is transmitted using AES-256 encryption and supports import. In step one, after establishing the communication link between the monitoring module and the flight control system, the GPS positioning module and terrain preprocessing data onboard the UAV are used to determine whether the monitoring area meets the collaborative triggering conditions. The triggering conditions are: when the monitoring area is ≥5km², or when the terrain preprocessing data shows monitoring blind spots with building obstructions and terrain elevation differences ≥30m, a multi-UAV collaborative monitoring mode is activated. During the dynamic correction process in step two, each UAV synchronizes flight control status data and dynamic interference correction coefficients in real time through distributed communication nodes. Then, the cross-machine collected data is calibrated a second time.

10. A drone-based human settlement environment monitoring system, applied to the drone-based human settlement environment monitoring method according to any one of claims 1-9, characterized in that: include: The monitoring module is used to set up a monitoring module and mount it on the UAV, and then establish a real-time communication link between the module and the flight control system. The monitoring module integrates gas sensing, visual imaging and attitude perception. The flight control unit is used to enable the UAV to fly over the human settlement environment. During the flight, the monitoring module collects parameters of the human settlement environment and then dynamically corrects the collected parameters of the monitoring module based on the flight control data to generate a calibrated data set. The scenario analysis module is used to input calibration data into the scenario-based analysis model, and then output quantitative indicators of potential hazards in the human living environment and present the visualization results. The early warning feedback module is used to trigger the flight control system to adjust the flight path or the ground terminal to provide early warning feedback based on the analysis results.

Images