An infrared unmanned aerial vehicle inspection method and system
The infrared drone inspection method, which automatically generates inspection paths and verifies multi-frame data, solves the problems of path planning mismatch and false fire alarms, achieves full-coverage inspection and accurate fire situation analysis, and improves inspection and rescue efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHAANXI CHANGYUKE AVIATION TECHNOLOGY CO LTD
- Filing Date
- 2026-04-03
- Publication Date
- 2026-07-03
AI Technical Summary
The lack of intelligent adaptation in the path planning of infrared drone inspections leads to incomplete coverage of the inspection area and a mismatch between the flight range and the drone's endurance. This results in insufficient fire early warning and emergency response capabilities, and the existing technology is susceptible to interference and false alarms, as well as insufficient rescue data support.
An automatic inspection path generation method is adopted, which generates a full-coverage path through a geographic model, combines multi-frame data verification and visible light-assisted analysis, transmits fire situation data in real time, and optimizes inspection paths and equipment adaptation.
It achieves comprehensive coverage of the inspection area, reduces false alarm rate, improves inspection and rescue efficiency, and provides accurate fire situation analysis support.
Smart Images
Figure CN122336901A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of inspection technology, and in particular to an infrared drone inspection method and system. Background Technology
[0002] Infrared drone inspection is an intelligent inspection solution that deeply integrates infrared thermal imaging technology with drone platforms. By using drones equipped with infrared sensors to fly autonomously, it can perform non-contact temperature detection, anomaly identification, and emergency response in target areas. It is widely used in fields such as power operation and maintenance, forest fire prevention, and industrial equipment monitoring. Its core advantage lies in breaking through the spatial limitations of manual inspection and achieving efficient, accurate, and all-weather hazard investigation.
[0003] However, there are two major pain points in the current field of infrared drone inspection. First, the path planning lacks intelligent adaptation. Traditional inspection paths rely heavily on manual experience to set, which can easily lead to incomplete coverage of the inspection area and mismatch between flight range and drone endurance, resulting in inspection interruptions or omissions of key areas. Second, the fire early warning and emergency response capabilities are insufficient. Existing technologies only judge the fire situation through a single infrared temperature measurement, which is easily affected by direct sunlight, normal equipment heating, and other interferences that can produce false alarms. Moreover, after a fire occurs, it cannot provide accurate situational data in real time, making it difficult to support fire rescue decision-making.
[0004] Therefore, an infrared drone inspection method and system are proposed to solve or alleviate the above problems. Summary of the Invention
[0005] The purpose of this invention is to provide an infrared unmanned aerial vehicle (UAV) inspection method and system to solve the problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: an infrared unmanned aerial vehicle (UAV) inspection method, comprising the following specific steps: Step 1: Collect infrared drone parameters, input geographical information of the inspection area, and automatically generate the inspection path; Step 2: Input the inspection path into the infrared drone to enable the infrared drone to perform infrared inspection; Step 3: Analyze the infrared drone inspection data in real time to determine whether the inspection location is a high-risk detection point, and verify it with refined inspection. When a fire is confirmed, send a fire alarm and assist personnel in extinguishing the fire. Step 4: After the infrared drone inspection is completed, record the inspection data and optimize the inspection path.
[0007] Preferably, in step one, inputting the geographic information of the inspection area involves inputting a coordinate range into the geographic model and calculating the selected area range. Based on the collected infrared UAV parameters, a preset inspection path is generated in the geographic model, ensuring that the infrared UAV can fully cover the inspection area when following the preset inspection path. Then, the preset inspection path is statistically analyzed to obtain a preset flight range. The preset flight range is compared with the infrared UAV's endurance. If the preset flight range is greater than the infrared UAV's endurance, the inspection area range is narrowed until the preset flight range is less than the infrared UAV's endurance, at which point the inspection path is output.
[0008] Preferably, generating a preset inspection path in the geographic model includes the following steps: S1.1: Determine the inspection starting point based on the geographical information of the infrared UAV take-off and landing point and the inspection area, so that the inspection starting point is located within the inspection area and the distance between the inspection starting point and the infrared UAV take-off and landing point is minimized. S1.2: Based on the fixed-point inspection range of the infrared UAV, the inspection area is divided into grids in the geographic model so that the inspection path of the infrared UAV maximizes the coverage of the grids within the inspection area and ensures that no grids within the inspection area are missed in adjacent paths. S1.3: The infrared drone starts by circling around the inspection starting point, performs reciprocating inspections of the area, and returns to the end point in a closed loop, ensuring that the inspection area is covered without blind spots and generating a preset inspection path.
[0009] Preferably, the infrared inspection performed by the infrared drone in step two includes the following steps: S2.1: Import the inspection path generated in step one into the flight control system of the infrared UAV through the wireless communication module, and make the flight control system of the infrared UAV automatically verify the integrity of the path data. S2.2: After completing path verification, the infrared UAV starts autonomous flight from the takeoff point according to the preset planned path, and controls the flight attitude of the infrared UAV to collect infrared data in the inspection area and transmit the collected data to the ground station in real time through wireless transmission technology. S2.3: When the infrared drone encounters an obstacle during flight, the infrared drone automatically activates the obstacle avoidance system, temporarily deviates from the original path and records the deviation node. After the risk is eliminated, it automatically returns to the preset route and ensures that the inspection coverage is not affected. S2.4: When a manual intervention command is received from the ground station during the journey, the UAV will execute the command first and update the remaining path simultaneously.
[0010] Preferably, when updating the remaining path in S2.4, the remaining range of the infrared UAV is first combined with the remaining path in the inspection path to determine whether the remaining range of the infrared UAV can complete the remaining inspection path. If the remaining inspection path can be completed, the UAV returns to the preset route for inspection. If the remaining inspection path cannot be completed, the remaining inspection path is shortened until the infrared UAV can return after completing the shortened remaining inspection path. If the infrared UAV cannot return autonomously after completing the manual intervention command from the ground station, it searches for a suitable landing point and sends the landing point geographical information to the ground station for subsequent retrieval of the infrared UAV.
[0011] Preferably, the high-risk detection point in step three is when the infrared drone's temperature exceeds a preset normal temperature threshold range during flight temperature measurement. Different normal temperature threshold ranges are set for different environmental types. When three consecutive frames of infrared data show a temperature point exceeding the threshold by more than 20%, it is marked as a suspected high-temperature point. Then, the infrared drone performs a refined inspection, reducing its flight altitude and speed, and repeatedly performs infrared temperature measurement on the target area from multiple angles. At the same time, it captures visible light images to assist in analysis, thus doubly determining whether a fire has occurred and eliminating non-fire-related high-temperature points.
[0012] Preferably, when a fire is determined to have occurred, the following operational steps are performed: S3.1: The drone activates its audible and visual alarm devices, hovers over the fire area to issue warnings, and sends fire information to the ground station and fire command center; S3.2: The drone adjusts to the optimal shooting angle, continuously records changes in flame temperature, smoke diffusion direction, and fire spread speed, activates the gas detection module, collects data such as carbon monoxide and oxygen concentration on site, automatically generates a fire situation analysis report, updates the data once every cycle, and synchronizes it to the fire platform to provide data support for fire fighting decisions. S3.3: When the infrared drone has sufficient power, it will continuously hover and monitor until firefighters arrive, assisting them in extinguishing the fire. When the infrared drone has insufficient power, it will return to the nearest alternate landing point and simultaneously transmit the stored local data back to the ground station.
[0013] Preferably, the inspection data recorded in step four includes a core detection data layer, an equipment performance data layer, and an abnormal event data layer. The core detection data layer is used to completely store the full frame data of infrared thermal imaging and associate the infrared data with the inspection path nodes one by one. The equipment performance data layer records the drone's full-process endurance consumption rate, infrared sensor frame rate stability, data transmission delay, obstacle avoidance trigger times, and calculates the deviation value between the actual flight distance and the preset path, as well as the infrared temperature measurement accuracy error, as equipment health assessment indicators. The abnormal event data layer establishes a closed-loop full-link record for event triggering, emergency handling, and results for high temperature points, fire alarms, and path deviation events.
[0014] Preferably, the optimization of the inspection path in step four includes coverage accuracy optimization, flight efficiency optimization, and equipment adaptation optimization. Coverage accuracy optimization involves comparing the actual coverage area of the inspection path with the preset area. For areas with blind spots, inspection nodes are added and flight path angles are adjusted. Fixed-point retest nodes are set in the path to increase the inspection frequency and dwell time in the area. Flight efficiency optimization involves analyzing the flight speed and turning time of infrared UAVs on different road sections, combining the flight endurance parameters, smoothing the route, removing redundant turning nodes, optimizing the altitude settings of long straight routes, and establishing charging stations in the inspection area. Based on the battery level of the infrared UAV, a charging station is selected for charging to improve the inspection range of the infrared UAV. Equipment adaptation optimization involves adjusting the inspection path of aging equipment based on the accuracy attenuation data of infrared sensors, and replacing the equipment. The accuracy compensation adjustment involves reducing the flight altitude and increasing the overlap rate of adjacent flight paths.
[0015] Another objective of this invention is to provide an infrared unmanned aerial vehicle (UAV) inspection system, comprising: The geographic information module constructs a geographic model of the inspection area, inputs coordinate range and infrared UAV parameters, and generates a preset inspection path with full coverage. The flight control module is used to receive the inspection path and import it into the UAV system to control the autonomous flight of the infrared UAV. The data acquisition module is used for collecting data during infrared drone inspections. A wireless communication module, which is used for bidirectional data transmission between the infrared UAV and the ground station; The intelligent analysis module is used to analyze infrared data in real time; Emergency response module, which is used to activate audible and visual alarms and send fire information to ground station and fire command center when a fire is detected; Ground station control module, which monitors the real-time status of the UAV; A data storage and management module is used to store and manage the data from infrared drone inspections. The path optimization module is used to optimize the coverage accuracy, flight efficiency, and equipment compatibility of infrared UAV inspections.
[0016] The technical effects and advantages of this invention are as follows: This invention utilizes an infrared UAV inspection method to generate a preset path based on the coordinates of the inspection area and the UAV parameters. It ensures that no adjacent paths are missed by dividing the path into grids. At the same time, it compares the preset flight range with the remaining flight range and automatically narrows the area until the flight range is suitable, avoiding inspection interruptions due to insufficient flight range. It adopts a path logic of starting from the outer perimeter, reciprocating inspection of the area, and returning to the endpoint in a closed loop, so as to achieve no dead-angle coverage of the inspection area and greatly improve the inspection efficiency and completeness. This invention solves the problems of high false alarm rate and insufficient rescue data support in traditional single temperature measurement by constructing a fire detection system with multi-frame data verification and visible light-assisted analysis, combined with real-time situation transmission function. Specifically, when infrared data shows a temperature point exceeding the threshold by 20% for three consecutive frames, the drone automatically lowers its altitude and slows down to conduct a detailed inspection, while simultaneously capturing visible light images for cross-verification, effectively identifying non-fire high-temperature points such as transformer overload and direct sunlight. After fire is confirmed, the drone transmits data such as flame temperature and smoke diffusion direction in real time, generating a dynamic situation analysis report that is synchronized to the fire platform, providing accurate basis for fire fighting decisions and significantly improving rescue efficiency. Attached Figure Description
[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the inspection method of the present invention; Figure 2 This invention provides an automatic inspection path diagram. Detailed Implementation
[0018] 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.
[0019] This invention provides, for example Figures 1-2 The infrared drone inspection method shown includes the following specific steps: Step 1: Collect infrared drone parameters, input geographical information of the inspection area, and automatically generate an inspection path so that the infrared drone can carry out flight inspections based on the automatically generated inspection path. Step 2: Input the inspection path into the infrared drone to enable the infrared drone to perform infrared inspection; Step 3: Analyze the infrared drone inspection data in real time to determine whether the inspection location is a high-risk detection point, and verify it with refined inspection. When a fire is confirmed, send a fire alarm and assist personnel in extinguishing the fire. Step 4: After the infrared drone inspection is completed, record the inspection data to facilitate subsequent data traceability, optimize the inspection path, and improve the efficiency and quality of the infrared drone inspection.
[0020] Furthermore, in step one, inputting the geographic information of the inspection area involves inputting the coordinate range into the geographic model and calculating the selected area range. Based on the collected infrared drone parameters (including, but not limited to, endurance parameters, infrared detection parameters, and shooting parameters), a preset inspection path is generated in the geographic model. A geographic model is a tool for abstracting, simplifying, and expressing geographic systems, phenomena, or processes. It aims to help people understand geographic laws, predict geographic changes, and solve geographic-related problems. It transforms complex geographic realities by extracting core elements and relationships. The process is transformed into an analyzable and simulable form, enabling the infrared UAV to cover the entire inspection area on the map when following the preset inspection path. Then, the preset inspection path is statistically analyzed to obtain the preset flight distance. The preset flight distance includes the round-trip distance of the infrared UAV from the take-off point to the inspection point and the flight distance within the inspection area. The preset flight distance is compared with the infrared UAV's endurance. When the preset flight distance is greater than the infrared UAV's endurance, the inspection area is narrowed down until the preset flight distance is less than the infrared UAV's endurance. Then, the inspection path is output, avoiding the situation where the infrared UAV cannot complete the inspection and has to return when the preset flight distance is greater than the infrared UAV's endurance.
[0021] Furthermore, generating a preset inspection path in the geographic model includes the following steps: S1.1: Determine the inspection starting point based on the geographical information of the infrared UAV take-off and landing point and the inspection area, so that the inspection starting point is located within the inspection area and the distance between the inspection starting point and the infrared UAV take-off and landing point is minimized. S1.2: Based on the fixed-point inspection range of the infrared UAV, the inspection area is divided into grids in the geographic model so that the inspection path of the infrared UAV can maximize the coverage of the grids within the inspection area and ensure that no grids within the inspection area are missed in adjacent paths, so as to automatically generate inspection paths based on the inspection area range. S1.3: The infrared drone initiates an outer perimeter circling inspection from the inspection starting point, performs a reciprocating area inspection, and returns to the closed loop endpoint. Starting from the starting point, it travels along the bottom horizontal channel to any end. Upon reaching the endpoint, it ascends along the vertical channel to the top, completing initial coverage of the area boundary. At the top horizontal channel, it turns and sequentially enters each vertical inspection area, performing a detailed inspection of all vertical areas in a reciprocating pattern from bottom to top. After completing the inspection of the last vertical area, it returns to the starting point along the top horizontal channel, ultimately returning to the inspection starting point, forming a complete inspection closed loop. This ensures comprehensive coverage of the inspection area and generates a preset inspection path, such as... Figure 2 As shown, the closed-loop inspection path can not only improve inspection efficiency, but also ensure inspection coverage.
[0022] Furthermore, step two, where the infrared drone performs infrared inspections, includes the following steps: S2.1: The inspection path generated in step one is imported into the flight control system of the infrared UAV through the wireless communication module, and the flight control system of the infrared UAV automatically verifies the integrity of the path data to avoid missing path data, which would prevent the infrared UAV from completing the inspection. In the event of a malicious network attack or signal interference during the flight inspection after the infrared UAV has completed path verification, resulting in missing path data, the final return command is executed to make the infrared UAV fly back and report the abnormal data. S2.2: After completing path verification, the infrared UAV starts autonomous flight from the takeoff point according to the preset planned path and controls the flight attitude of the infrared UAV to collect infrared data in the inspection area and transmit the collected data to the ground station in real time via wireless transmission technology so that the ground station can collect and analyze the window data. S2.3: When the infrared drone encounters an obstacle during flight, the infrared drone automatically activates the obstacle avoidance system, temporarily deviates from the original path and records the deviation node. After the risk is eliminated, it automatically returns to the preset route and ensures that the inspection coverage is not affected, so as to ensure the normal flight inspection of the infrared drone. S2.4: After completing the inspection of each grid area, the UAV sends an area completion signal to the ground station. The ground system automatically marks the inspected area. If it receives manual intervention instructions from the ground station during the process, such as adding a new inspection point or temporarily returning to base, the UAV will execute the instructions first and update the remaining path simultaneously, thereby ensuring the inspection of subsequent areas.
[0023] Furthermore, when updating the remaining path in S2.4, it first combines the remaining range of the infrared UAV with the remaining path in the inspection path to determine whether the infrared UAV's remaining range can complete the remaining inspection path. If it can complete the remaining inspection path, it returns to the preset route for inspection. If it cannot complete the remaining inspection path, it shortens the remaining inspection path until the infrared UAV can return after completing the shortened remaining inspection path. This reduces the possibility of the infrared UAV being unable to return and forced to land, thus reducing the need for subsequent personnel recovery. Moreover, if the infrared UAV cannot return autonomously after completing the manual intervention command from the ground station, it searches for a suitable landing point and sends the landing point's geographical information to the ground station for subsequent retrieval of the infrared UAV. This reduces the possibility of the infrared UAV crashing due to power failure and reduces the waste of equipment resources.
[0024] In particular, the high-risk detection points in step three are those where the infrared drone's temperature exceeds the preset normal temperature threshold range during flight temperature measurement. Different normal temperature threshold ranges are set for different environmental types, such as transformers, cable joints, and building walls. The normal operating temperature of transformers is ≤85℃, and that of cable joints is ≤70℃. When three consecutive frames of infrared data show temperatures exceeding the threshold by more than 20%, these points are marked as suspected high-temperature points. The infrared drone then conducts a refined inspection, reducing its flight altitude and speed, and repeatedly performing infrared temperature measurements on the target area from multiple angles. Simultaneously, it captures visible light images for auxiliary analysis, doubly confirming whether a fire has occurred and investigating non-fire-related high-temperature points. The investigation methods include multi-dimensional data cross-validation, environmental factor investigation, and manual verification. Cross-validation of dimensional data involves comparing historical temperature data for a given point. If the temperature is normal and the equipment is heating up (e.g., transformer overload, motor operation), and the temperature remains stable within a safe range, it is marked as a normal heating point, thus eliminating it as a suspected high-temperature point. Environmental factor investigation involves using visible light images to determine if there is direct sunlight, blocked equipment vents, or external heat sources such as welding operations. If the temperature anomaly is confirmed to be caused by external interference, the interference source information is recorded and the warning is lifted. Manual verification involves the platform pushing thermal images, visible light images, and temperature curves to the inspection personnel's terminals for manual confirmation of whether it is a non-fire hazard. If there is no response from the ground station, a fire alarm is issued, and the point is continuously monitored, with monitoring data transmitted in real time.
[0025] Furthermore, when a fire is confirmed to have occurred, the following operational procedures are followed: S3.1: The drone activates its audible and visual alarm devices, hovers over the fire area to warn people, and sends fire information to the ground station and fire command center, thereby alerting people and animals in the area so that they can respond to the fire. S3.2: The drone adjusts to the optimal shooting angle and continuously records changes in flame temperature, smoke diffusion direction, and fire spread speed. When the infrared drone is equipped with a gas detection module, it starts the gas detection module to collect data such as carbon monoxide and oxygen concentration on site, automatically generates a fire situation analysis report, updates the data once every one cycle, and synchronizes it to the fire platform to provide data support for fire fighting decisions, thereby improving fire fighting efficiency and reducing damage and waste of property resources. S3.3: When the infrared drone has sufficient power, it will continuously hover and monitor until firefighters arrive, assisting them in extinguishing the fire. When the infrared drone has insufficient power, it will return to the nearest alternate landing point and simultaneously transmit the stored local data back to the ground station, thus ensuring the protection of the infrared drone in case of loss or damage.
[0026] Furthermore, the inspection data recorded in step four includes a core detection data layer, an equipment performance data layer, and an anomaly event data layer. The core detection data layer is used to completely store full-frame infrared thermal imaging data, visible light images, and flight trajectory logs, including the temperature value, temperature measurement timestamp, and latitude and longitude coordinates of each pixel, with an accuracy of ≤0.5 meters. It also associates the infrared data with each node of the inspection path, similar to the correspondence between product images and standard parameters in battery testing. The equipment performance data layer records parameters such as the drone's overall endurance consumption rate, infrared sensor frame rate stability, data transmission latency, and obstacle avoidance trigger count, while also calculating the actual flight distance and the preset distance. The deviation value of the path and the accuracy error of infrared temperature measurement, compared with the data of ground calibration equipment, serve as indicators for equipment health assessment. The abnormal event data layer establishes a closed-loop record of the entire chain of events, including high temperature points, fire alarms, and path deviation events, from event triggering and emergency handling to results. It includes temperature change curves of abnormal points, manual review conclusions, emergency response actions, and event handling time. It is analogous to the traceability system for non-conforming products, error analysis, and conclusion determination in battery testing. All data is transmitted to the cloud database through encryption, generating a unique inspection task ID. It supports multi-dimensional retrieval by time, region, and equipment type, providing a traceable data source for subsequent path optimization.
[0027] In particular, step four, which optimizes the inspection route, includes coverage accuracy optimization, flight efficiency optimization, and equipment compatibility optimization. Coverage accuracy optimization involves comparing the actual coverage area of the inspection route with the preset area. For areas with blind spots, inspection nodes are added and flight path angles are adjusted. The methods for adding inspection nodes include automatic addition and manual addition. Manual addition involves staff adding inspection points for key protected equipment and areas. Automatic addition involves adding inspection points based on high-risk detection points in historical inspection data and statistically analyzing the distribution patterns of high-frequency anomalies. For example, if a cable joint experiences three high-temperature warnings per month, a fixed-point retest node is set in the route to increase the inspection frequency and dwell time in that area. Flight efficiency optimization involves analyzing the flight speed and turning time of infrared UAVs on different road sections, combining this with endurance parameters to smooth the path, remove redundant turning nodes, and optimize the altitude settings for long straight routes. For example, increasing the flight altitude in plain areas from 50 meters to 80 meters can shorten the total flight range by 10%-15%. This is analogous to the mechanism of adjusting the platform height in battery testing to optimize imaging efficiency. Furthermore, charging stations are established within the inspection area, and charging is performed at selected stations based on the infrared UAV's battery level. This increases the inspection range of the infrared UAV, thereby expanding the inspection area and improving its inspection efficiency. Equipment adaptation optimization involves adjusting the inspection paths of aging equipment based on the accuracy attenuation data of infrared sensors, and replacing the equipment. Accuracy compensation adjustment involves lowering the flight altitude and increasing the overlap rate of adjacent routes to ensure that the detection accuracy meets the requirements. This is similar to the approach of updating calibration parameters in battery testing to adapt to equipment aging. Combined with the performance parameters of different UAV models, such as endurance, detection range, and obstacle avoidance capabilities, an equipment and path matching model is generated. This automatically matches suitable inspection areas and path planning schemes for new equipment. For example, it matches cross-regional inspection paths for long-endurance UAVs and fine-grained inspection paths for high-resolution infrared UAVs. Equipment adaptation optimization also includes maintenance requirement assessment, equipment replacement decision assessment, and equipment service life assessment. The maintenance needs assessment is tiered, and maintenance priorities are determined. Tiered maintenance standards include daily maintenance, quarterly in-depth maintenance, and annual calibration maintenance. Daily maintenance involves basic accuracy calibration of infrared sensors, checking the path deviation rate of the UAV flight path planning system, cleaning dust from sensor lenses, and recording battery capacity decay rate. Quarterly in-depth maintenance involves comprehensive testing of the pixel response consistency of infrared sensors, wear detection of the UAV power system, updating the environmental recognition parameters of the obstacle avoidance system, and adjusting the compensation benchmark value of flight altitude and flight path overlap rate based on accuracy decay data. Annual calibration maintenance involves collaborating with a third-party organization to conduct authoritative calibration of infrared sensor accuracy, comparing the test data before and after calibration, updating the accuracy compensation model parameters, and assessing whether the overall performance of the UAV meets the inspection requirements. Maintenance priority is determined by establishing a two-dimensional evaluation matrix of accuracy decay rate and equipment failure frequency. Equipment with a monthly accuracy decay rate exceeding 5% and a monthly failure frequency ≥ 2 times is classified as Level 1 maintenance priority and must be maintained within 3 working days. Equipment with an accuracy decay rate between 2% and 5% and no high-frequency failures is classified as Level 2 maintenance priority and is executed according to the quarterly maintenance plan. Equipment replacement decision assessment includes replacement triggering conditions and replacement scheme matching. Replacement triggering conditions include accuracy limit threshold, cost-effectiveness threshold, and performance iteration threshold. Accuracy limit threshold is triggered when the accuracy of the infrared sensor decays to less than 60% of its initial value, and after accuracy compensation measures such as reducing flight altitude and increasing flight path overlap, the detection accuracy still cannot meet industry standards, thus triggering mandatory replacement. Cost-effectiveness threshold is triggered when the annual maintenance cost of a single device exceeds 60% of the purchase cost of new equipment, or when the accuracy decay rate is not effectively controlled after three consecutive maintenance cycles, thus initiating a replacement assessment. Performance iteration threshold is triggered when the new equipment improves by more than 30% in core parameters such as infrared resolution, endurance, and obstacle avoidance efficiency compared to the existing equipment. Furthermore, it can cover more complex inspection scenarios, such as cross-regional inspections in mountainous areas. Batch replacement plans can be initiated in advance. The replacement scheme matching is based on the equipment and path matching model to match the optimal replacement model for the equipment to be replaced. For example, after the original long-endurance drone reaches the end of its service life, a new model with a 25% increase in endurance and multi-sensor fusion detection capability will be prioritized to continue cross-regional inspection tasks. When replacing the original high-resolution infrared drone, a sensor model with a 40% increase in resolution will be selected to further enhance the detection accuracy in refined inspection scenarios. A batch replacement plan will be formulated to prioritize the replacement of old equipment that has reached the end of its service life and has high maintenance costs, while retaining some high-performance equipment as a backup to ensure the continuity of inspection tasks. Equipment service life assessment includes sensor lifespan threshold setting and a dynamic lifespan correction mechanism. The sensor lifespan threshold setting is based on historical data of infrared sensor accuracy decay, establishing a correlation model between accuracy decay rate and service duration. A critical service life is set when sensor accuracy decays to 70% of its initial value. This threshold can be adjusted according to industry testing standards. Combined with the design lifespan of core UAV components such as the power system and flight control system, the maximum service life of the entire aircraft is determined, such as 8 years for fixed-wing UAVs and 5 years for multi-rotor UAVs. Simultaneously, the actual service time of the equipment is recorded, and a retirement warning period is initiated 6 months in advance. The dynamic lifespan correction mechanism dynamically adjusts the remaining service life based on the accelerated impact of the actual inspection environment on accuracy decay. For example, equipment operating in coastal salt spray environments experiences a 20% increase in accuracy decay rate, resulting in a 15% reduction in service life. The lifespan prediction model is updated based on the accuracy recovery after each maintenance, avoiding over-service or premature retirement that would waste resources.
[0028] Another objective of this invention is to provide an infrared drone inspection system, comprising a geographic information module, a flight control module, a data acquisition module, a wireless communication module, an intelligent analysis module, an emergency response module, a ground station management module, a data storage and management module, and a path optimization module. The geographic information module constructs a geographic model of the inspection area, inputs coordinate ranges and infrared drone parameters, generates a fully covered preset inspection path, verifies the matching degree between the preset flight range and the infrared drone's endurance, and dynamically adjusts the inspection area range. The flight control module receives the inspection path and imports it into the drone system, controls the infrared drone to fly autonomously, adjusts its flight attitude to adapt to infrared data acquisition requirements, handles manual intervention commands such as temporary return to base, adding detection points, and synchronously updates the remaining path. The data acquisition module collects data from the infrared drone inspection, including the temperature value, temperature measurement timestamp, and latitude and longitude coordinates of each pixel, and synchronously acquires visible light images for auxiliary analysis. The wireless communication module transmits data bidirectionally between the infrared drone and the ground station, including path command issuance, real-time data feedback, and manual commands. The system includes features such as receiving and sending fire alarms, supporting encrypted transmission to a cloud database. The intelligent analysis module analyzes infrared data in real time, identifies high-temperature anomalies, combines multi-frame data with visible light images to assess fire risk, and generates fire situation analysis reports. The emergency response module activates audible and visual alarms upon fire detection, sends fire information to the ground station and fire command center, continuously hovers to monitor the fire situation, and provides data support for firefighting decisions. It automatically plans alternative landing paths when battery is low. The ground station control module monitors the drone's real-time status, marks inspected areas, sends manual intervention commands, receives, stores, and manages all inspection data, generates inspection task reports, and supports multi-dimensional retrieval. The data storage and management module stores and manages the infrared drone's inspection data. The path optimization module optimizes the coverage accuracy, flight efficiency, and equipment compatibility of the infrared drone's inspections. The system is interconnected with the geographic information module, flight control module, data acquisition module, wireless communication module, intelligent analysis module, emergency response module, ground station control module, data storage and management module, and path optimization module.
[0029] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. An infrared unmanned aerial vehicle (UAV) inspection method, characterized in that: The specific steps include the following: Step 1: Collect infrared drone parameters, input geographical information of the inspection area, and automatically generate the inspection path; Step 2: Input the inspection path into the infrared drone to enable the infrared drone to perform infrared inspection; Step 3: Analyze the infrared drone inspection data in real time to determine whether the inspection location is a high-risk detection point, and verify it with refined inspection. When a fire is confirmed, send a fire alarm and assist personnel in extinguishing the fire. Step 4: After the infrared drone inspection is completed, record the inspection data and optimize the inspection path.
2. The method of claim 1, wherein: In step one, inputting the geographic information of the inspection area involves inputting a coordinate range into the geographic model and calculating the selected area range. Based on the collected infrared UAV parameters, a preset inspection path is generated in the geographic model, ensuring that the infrared UAV can fully cover the inspection area when following the preset inspection path. Then, the preset inspection path is statistically analyzed to obtain the preset flight range. The preset flight range is compared with the infrared UAV's endurance. If the preset flight range is greater than the infrared UAV's endurance, the inspection area range is narrowed until the preset flight range is less than the infrared UAV's endurance, at which point the inspection path is output.
3. The method of claim 2, wherein: The generation of the preset inspection path in the geographic model includes the following steps: S1.1: Determine the inspection starting point based on the geographical information of the infrared UAV take-off and landing point and the inspection area, so that the inspection starting point is located within the inspection area and the distance between the inspection starting point and the infrared UAV take-off and landing point is minimized. S1.2: Based on the fixed-point inspection range of the infrared UAV, the inspection area is divided into grids in the geographic model so that the inspection path of the infrared UAV maximizes the coverage of the grids within the inspection area and ensures that no grids within the inspection area are missed in adjacent paths. S1.3: The infrared drone starts by circling around the inspection starting point, performs reciprocating inspections of the area, and returns to the end point in a closed loop, ensuring that the inspection area is covered without blind spots and generating a preset inspection path.
4. The infrared UAV inspection method according to claim 1, characterized in that: Step two, where the infrared drone performs infrared inspection, includes the following steps: S2.1: Import the inspection path generated in step one into the flight control system of the infrared UAV through the wireless communication module, and make the flight control system of the infrared UAV automatically verify the integrity of the path data. S2.2: After completing path verification, the infrared UAV starts autonomous flight from the takeoff point according to the preset planned path, and controls the flight attitude of the infrared UAV to collect infrared data in the inspection area and transmit the collected data to the ground station in real time through wireless transmission technology. S2.3: When the infrared drone encounters an obstacle during flight, the infrared drone automatically activates the obstacle avoidance system, temporarily deviates from the original path and records the deviation node. After the risk is eliminated, it automatically returns to the preset route and ensures that the inspection coverage is not affected. S2.4: When a manual intervention command is received from the ground station during the journey, the UAV will execute the command first and update the remaining path simultaneously.
5. The infrared UAV inspection method according to claim 4, characterized in that: When updating the remaining path in S2.4, the remaining range of the infrared UAV is first combined with the remaining path in the inspection path to determine whether the remaining range of the infrared UAV can complete the remaining inspection path. If the remaining inspection path can be completed, the UAV returns to the preset route for inspection. If the remaining inspection path cannot be completed, the remaining inspection path is shortened until the infrared UAV can return after completing the shortened remaining inspection path. If the infrared UAV cannot return autonomously after completing the manual intervention command from the ground station, it searches for a suitable landing point and sends the landing point's geographical information to the ground station for subsequent retrieval of the infrared UAV.
6. The infrared UAV inspection method according to claim 1, characterized in that: The high-risk detection point in step three is when the infrared drone's temperature exceeds the preset normal temperature threshold range during flight temperature measurement. Different normal temperature threshold ranges are set for different environmental types. When three consecutive frames of infrared data show a temperature point exceeding the threshold by more than 20%, it is marked as a suspected high-temperature point. Then, the infrared drone performs a refined inspection, reducing its flight altitude and speed, and repeatedly performs infrared temperature measurement on the target area from multiple angles. At the same time, it captures visible light images to assist in analysis, thus doubly determining whether a fire has occurred and eliminating non-fire-related high-temperature points.
7. The infrared UAV inspection method according to claim 6, characterized in that: When a fire is confirmed to have occurred, the following procedures shall be followed: S3.1: The drone activates its audible and visual alarm devices, hovers over the fire area to issue warnings, and sends fire information to the ground station and fire command center; S3.2: The drone adjusts to the optimal shooting angle, continuously records changes in flame temperature, smoke diffusion direction, and fire spread speed, activates the gas detection module, collects data such as carbon monoxide and oxygen concentration on site, automatically generates a fire situation analysis report, updates the data once every cycle, and synchronizes it to the fire platform to provide data support for fire fighting decisions. S3.3: When the infrared drone has sufficient power, it will continuously hover and monitor until firefighters arrive, assisting them in extinguishing the fire. When the infrared drone has insufficient power, it will return to the nearest alternate landing point and simultaneously transmit the stored local data back to the ground station.
8. The infrared UAV inspection method according to claim 1, characterized in that: The inspection data recorded in step four includes a core detection data layer, an equipment performance data layer, and an abnormal event data layer. The core detection data layer is used to completely store the full frame data of infrared thermal imaging and associate the infrared data with the inspection path nodes one by one. The equipment performance data layer records the drone's endurance consumption rate, infrared sensor frame rate stability, data transmission latency, and obstacle avoidance trigger count parameters. It also calculates the deviation between the actual flight distance and the preset path, as well as the infrared temperature measurement accuracy error, as indicators for equipment health assessment. The abnormal event data layer establishes a closed-loop record of event triggering, emergency handling, and results for high temperature points, fire alarms, and path deviation events.
9. The infrared UAV inspection method according to claim 8, characterized in that: The fourth step of optimizing the inspection path includes coverage accuracy optimization, flight efficiency optimization, and equipment adaptation optimization. Coverage accuracy optimization involves comparing the actual coverage area of the inspection path with the preset area. For areas with blind spots, inspection nodes are added and flight path angles are adjusted. Fixed-point retest nodes are set in the path to increase the inspection frequency and dwell time in the area. Flight efficiency optimization involves analyzing the flight speed and turning time of infrared UAVs on different road sections, combining the flight endurance parameters, smoothing the route, removing redundant turning nodes, optimizing the altitude settings of long straight routes, and establishing charging stations in the inspection area. Based on the battery level of the infrared UAV, a charging station is selected for charging to improve the inspection range of the infrared UAV. Equipment adaptation optimization involves adjusting the inspection path of aging equipment based on the accuracy attenuation data of infrared sensors, and replacing the equipment. The accuracy compensation adjustment involves reducing the flight altitude and increasing the overlap rate of adjacent flight paths.
10. An infrared drone inspection system, implementing the infrared drone inspection method according to any one of claims 1-9, characterized in that: include: The geographic information module constructs a geographic model of the inspection area, inputs coordinate range and infrared UAV parameters, and generates a preset inspection path with full coverage. The flight control module is used to receive the inspection path and import it into the UAV system to control the autonomous flight of the infrared UAV. The data acquisition module is used for collecting data during infrared drone inspections. A wireless communication module, which is used for bidirectional data transmission between the infrared UAV and the ground station; The intelligent analysis module is used to analyze infrared data in real time; Emergency response module, which is used to activate audible and visual alarms and send fire information to ground station and fire command center when a fire is detected; Ground station control module, which monitors the real-time status of the UAV; A data storage and management module is used to store and manage the data from infrared drone inspections. The path optimization module is used to optimize the coverage accuracy, flight efficiency, and equipment compatibility of infrared UAV inspections.