Unmanned aerial vehicle infrared thermal imaging-based ship battery detection system and method

Abstract

This invention discloses a ship battery inspection system and method based on UAV infrared thermal imaging. The system includes: a UAV for autonomously flying inside the ship's battery compartment to collect infrared and visible light images; a battery parameter synchronous acquisition module for synchronously acquiring electrical parameters and spatiotemporally correlating them with infrared data; a data fusion and analysis module for preprocessing, thermal anomaly identification, and calculation of the battery health index (BHI); and a cloud-based health management module for data archiving and trend analysis. This invention enables contactless, high-efficiency, and comprehensive inspection of ship batteries, achieving early warning of thermal anomalies and quantitative assessment of health status, reducing inspection safety risks, ensuring ship navigation safety, and solving the problems of low efficiency, incomplete coverage, high safety risks, and insufficient risk prediction in traditional ship battery inspection methods.

Description

Technical Field

[0001] This invention relates to the field of marine electrical equipment testing and UAV application technology, specifically to a marine battery testing system and method based on UAV infrared thermal imaging. Background Technology

[0002] As a core component of a ship's power system and emergency support system, the operational status of marine batteries directly affects the safety of ship navigation. With the electrification and intelligent upgrading of ships, the application of large-capacity, high-density battery packs on ships is becoming increasingly widespread. However, during charge-discharge cycles and long-term use, batteries are prone to problems such as localized overheating, poor contact, abnormal internal resistance, and signs of thermal runaway. If these problems are not detected in time, they may lead to safety accidents such as fires and explosions.

[0003] Traditional ship battery testing mainly relies on manual point-by-point testing using handheld infrared thermometers or multimeters, which presents significant technical bottlenecks. (1) Low detection efficiency and incomplete coverage: Ship battery packs are usually arranged in small and enclosed spaces such as engine room and battery compartment. Manual detection requires disassembling protective devices and measuring the temperature and voltage of each battery. The detection time for a single battery pack can be as long as 1-2 hours, and thermal anomalies in hidden parts such as the battery pack interior and wiring terminals are easily missed.

[0004] (2) Delayed thermal anomaly identification and insufficient risk prediction: Manual detection can only obtain instantaneous temperature data, which cannot capture the dynamic trend of battery temperature changes, making it difficult to identify early minor thermal anomalies (such as local temperature rise of 3-5℃), and cannot be associated with factors such as battery charging and discharging status, ambient temperature and humidity to make risk prediction, which may lead to missing the best time for handling.

[0005] (3) High operational safety risks: The battery compartment is small and poorly ventilated, posing risks such as electrolyte leakage and electromagnetic interference. Manual close-range inspection is prone to safety accidents such as electric shock and chemical corrosion. In particular, during ship navigation, the swaying of the ship further increases the difficulty and safety risks of manual inspection.

[0006] (4) Lack of systematic data management: Manual test data is mostly stored in paper records or simple spreadsheets, without unified database management, making it impossible to trace historical data, analyze trends and compare batches, and support the health management of the entire battery life cycle.

[0007] Existing UAV infrared detection technology is mostly used in fields such as construction and power, and its adaptability to ship battery detection is insufficient: it lacks the precise flight control capability in the confined space of ship battery compartments, the infrared thermal imaging data and battery parameters (voltage, internal resistance, charge and discharge status) are not correlated and analyzed, and there is no specific standard for ship batteries to determine thermal anomalies, which cannot meet the professional and standardized needs of ship battery detection. Summary of the Invention

[0008] The purpose of this invention is to propose a ship battery detection system and method based on UAV infrared thermal imaging. Through precise acquisition by UAV infrared thermal imaging, correlation analysis of battery parameters, and intelligent identification of thermal anomalies, it can achieve non-contact, high-efficiency, and all-round detection of ship batteries, realize early warning of thermal anomalies and quantitative assessment of health status, reduce detection safety risks, and ensure the safety of ship navigation. This solves the problems of low efficiency, incomplete coverage, high safety risks, and insufficient risk prediction in traditional ship battery detection.

[0009] To achieve the above objectives, in a first aspect, the present invention proposes a ship battery detection system based on UAV infrared thermal imaging, comprising: The drone is used to fly autonomously inside the ship's battery compartment. The drone is equipped with an infrared thermal imager and a visible light camera to collect infrared thermal imaging data and visible light images of the battery pack, and sends the infrared thermal imaging data to a data fusion analysis module. The battery parameter synchronous acquisition module is used to synchronously acquire the battery's electrical parameters through the battery management system interface or wireless sensor node, and to perform spatiotemporal correlation between the electrical parameters and the infrared thermal imaging data acquired by the UAV to obtain spatiotemporally correlated multi-source data, which is then sent to the data fusion analysis module. The data fusion and analysis module is used to receive the infrared thermal imaging data and the spatiotemporally correlated multi-source data, preprocess the infrared image, locate the abnormal area through the thermal anomaly identification algorithm, and calculate the battery health index (BHI) in combination with the electrical parameters to trigger a graded early warning. The cloud-based health management module is used to receive the test results output by the data fusion and analysis module, establish a ship battery test database, and realize data archiving, trend analysis, and multi-terminal access.

[0010] Secondly, this invention proposes a method for detecting ship batteries based on UAV infrared thermal imaging, implemented using the system described in the first aspect, and comprising the following steps: Step 1: Import the structural diagram of the target ship's battery compartment or select a ship type template on the ground terminal, set the inspection area, waypoint density and safety distance, generate the inspection route, and configure the BMS interface parameters; Step 2: The UAV flies along the planned inspection route, collecting infrared thermal imaging data and visible light images in real time; the battery parameter synchronous acquisition module collects battery electrical parameters and environmental data synchronously through the BMS interface or wireless sensor node, and completes spatiotemporal correlation annotation; Step 3: The data fusion analysis module receives the infrared thermal imaging data and the spatiotemporally correlated multi-source data, preprocesses the infrared image, locates the abnormal area through the thermal anomaly identification algorithm, determines the anomaly type by combining electrical parameters, and calculates the battery health index (BHI). Step 4: The data fusion and analysis module triggers graded early warnings based on the BHI value and / or local temperature and temperature rise rate, and automatically generates a standardized test report; Step 5: The ground terminal uploads the detection data to the cloud database, and the cloud health management module regularly generates a summary report on the battery health status.

[0011] The beneficial effects of this invention are as follows: 1. Achieving contactless and efficient inspection through a design adapted to the confined space of ship battery compartments. This invention optimizes the flight control and obstacle avoidance mechanisms of drones for the unique, enclosed, and confined environment of ship battery compartments. This allows drones to autonomously enter the compartment for operation, eliminating the need for manual entry into hazardous areas. This reduces the inspection time for a single battery group to less than 10 minutes, increasing inspection efficiency by more than 80% (6-12 times) compared to manual inspection, while completely eliminating safety risks such as electric shock and chemical corrosion.

[0012] 2. By employing multi-source data spatiotemporal correlation technology, the accuracy of anomaly detection is significantly improved. This invention establishes a precise correlation between infrared temperature data and battery electrical parameters and spatial location, effectively distinguishing between "normal temperature rise" and "fault-induced overheating," achieving a thermal anomaly identification accuracy rate of over 95%. It can identify early, minor thermal anomalies (temperature rise of 3-5°C), detecting potential faults 3-5 charge-discharge cycles earlier than manual detection, thus preventing thermal runaway accidents.

[0013] 3. A dedicated ship-specific thermal anomaly identification and health assessment system enables quantitative status monitoring and tiered early warning. Based on a ship battery fault characteristic library (containing 6 typical fault types) and a three-dimensional health assessment model (temperature + electrical + lifespan), this invention calculates the Battery Health Index (BHI) and automatically issues tiered early warnings, providing targeted handling suggestions. This represents a leap from "qualitative judgment" to "quantitative assessment," facilitating rapid decision-making by maintenance personnel. Simultaneously, it establishes a full lifecycle monitoring record, supporting trend analysis and prediction, which can extend battery lifespan by 10%-15% and reduce the risk of ship downtime.

[0014] 4. Enhanced Standardization and Traceability of Testing Through Intelligent Closed-Loop Management Throughout the Entire Process. This invention integrates functions such as autonomous flight planning, automatic data fusion and analysis, one-click report generation, cloud archiving, and multi-terminal access, forming a closed-loop management system of "detection-analysis-early warning-archiving-trend tracking." This standardizes the testing process, ensures data traceability, supports remote monitoring and batch comparison, significantly reduces reliance on human experience, and meets the urgent need for authoritative testing data in scenarios such as ship insurance underwriting and maritime spot checks.

[0015] The present invention has other features and advantages, which will be apparent from or will be set forth in detail in the accompanying drawings and the following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0016] The above and other objects, features and advantages of the present invention will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0017] Figure 1 A schematic diagram of an infrared ship battery detection system for unmanned aerial vehicles (UAVs) according to the present invention is shown. Detailed Implementation

[0018] The invention will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0019] Example 1

[0020] like Figure 1 As shown, the present invention provides an infrared ship battery detection system for unmanned aerial vehicles (UAVs), which adopts a three-layer architecture of "UAV infrared acquisition + ground terminal analysis + cloud data management". Specifically, it includes: a UAV, a battery parameter synchronous acquisition module, a data fusion analysis module, and a cloud health management module. The battery parameter synchronous acquisition module and the data fusion analysis module run on a ground terminal (such as a portable industrial tablet or ground station), while the cloud health management module is deployed on a cloud server.

[0021] The drone is used to fly autonomously inside the ship's battery compartment. The drone is equipped with an infrared thermal imager and a visible light camera to collect infrared thermal imaging data and visible light images of the battery pack, and transmits the infrared thermal imaging data back to the ground terminal in real time for the data fusion and analysis module.

[0022] In a preferred embodiment, the infrared thermal imager mounted on the UAV has a temperature measurement range of -20℃ to 150℃, a measurement accuracy of ±2%, and a temperature resolution of 0.1℃. The UAV is also equipped with an ultrasonic obstacle avoidance sensor and a visual obstacle avoidance module, maintaining a minimum safe distance of ≥100cm from the battery pack. The UAV's detection time for a single battery pack is ≤10 minutes, and the data transmission latency is ≤200ms. The UAV has a built-in digital model library of ship battery compartments, supports customizable detection areas and waypoint densities (default 5-10cm / point), and can achieve one-click takeoff, autonomous inspection, and automatic return. Data is transmitted in real-time via 4G image transmission and has a resume function.

[0023] The battery parameter synchronous acquisition module is used to synchronously acquire the battery's electrical parameters through the battery management system (BMS) interface or wireless sensor node, and to perform spatiotemporal correlation between the electrical parameters and the infrared thermal imaging data acquired by the UAV to obtain spatiotemporally correlated multi-source data, which is then sent to the data fusion analysis module.

[0024] In a preferred embodiment, the battery parameter synchronous acquisition module collects electrical parameters including the total battery pack voltage, individual cell voltage, charging / discharging current, internal resistance, and cycle count, with an acquisition frequency ≥1Hz. The spatiotemporal correlation refers to: assigning a unified timestamp to the infrared thermal imaging data and electrical parameters, with a timestamp synchronization error ≤100ms, and establishing a three-dimensional correlated dataset including temperature distribution, electrical parameters, and spatial location by combining it with UAV positioning data. This module also synchronously collects temperature and humidity data and ventilation status data within the battery compartment as an environmental correction basis for thermal anomaly detection.

[0025] In this embodiment, the data fusion analysis module is used to receive the infrared thermal imaging data and the spatiotemporally correlated multi-source data, preprocess the infrared image, locate the abnormal area through the thermal anomaly identification algorithm, and calculate the battery health index (BHI) in combination with the electrical parameters to trigger a graded early warning.

[0026] In a preferred embodiment, the data fusion analysis module further includes an environmental parameter adaptation unit for synchronously collecting temperature and humidity data and ventilation status data within the battery compartment, serving as the environmental correction basis for thermal anomaly determination. The thermal anomaly identification algorithm is based on a marine battery thermal anomaly feature library, which includes six typical fault types, including localized overheating, terminal heating, and poor battery consistency. The data fusion analysis module automatically labels the anomaly location, the highest temperature at the anomaly location, and the rate of temperature rise using the thermal anomaly identification algorithm, achieving a thermal anomaly identification accuracy of ≥95%.

[0027] The Battery Health Index (BHI) is calculated using the following formula: BHI = 0.4×Ts + 0.3×Ep + 0.3×Lc Wherein, Ts is the temperature state coefficient, ranging from 0 to 100, calculated based on the degree of thermal anomaly and temperature consistency; Ep is the electrical parameter coefficient, ranging from 0 to 100, calculated based on voltage consistency and internal resistance change rate; Lc is the lifespan coefficient, ranging from 0 to 100, calculated based on the ratio of cycle count to design life. The data fusion analysis module classifies data according to the BHI value: BHI ≥ 80 indicates a healthy state, 60 ≤ BHI < 80 indicates a state of concern, and BHI < 60 indicates a state of warning.

[0028] The cloud-based health management module is used to receive the test results output by the data fusion and analysis module, establish a ship battery test database, and realize data archiving, trend analysis, and multi-terminal access.

[0029] In a preferred embodiment, the cloud-based health management module establishes a ship battery testing database with a data retention period of ≥5 years, supporting multi-dimensional searches by ship name, battery pack number, or testing time. The cloud-based health management module automatically generates temperature change curves and health index decay trend charts for individual battery packs, triggering an early warning when the temperature rise rate is ≥0.5℃ / charge-discharge cycle. The cloud-based health management module supports login via computer, tablet, and mobile phone terminals, and supports remote issuance of testing tasks.

[0030] Through the above three-layer architecture, this invention realizes intelligent detection throughout the entire process from data acquisition and on-site analysis to cloud management, which not only ensures low latency for real-time response, but also meets the needs of long-term data management and remote access.

[0031] Example 2

[0032] Based on the system of Embodiment 1 above, this embodiment provides a method for detecting infrared ship batteries using a drone, including the following steps: Step 1: Testing Preparation and Planning Import the structural diagram of the target ship's battery compartment or select a ship type template on the ground terminal, set the detection area, waypoint density (default 5-10cm / point) and safety distance, generate the inspection route, and configure the BMS interface parameters.

[0033] Step 2: Autonomous Detection and Data Acquisition by the UAV

[0034] The UAV flies along a planned route, collecting infrared thermal imaging data and visible light images in real time. The battery parameter synchronous acquisition module collects battery electrical parameters and environmental data synchronously via the BMS interface or wireless sensor nodes, and performs spatiotemporal correlation annotation (applying a unified timestamp to the infrared data and battery parameters, and establishing a three-dimensional correlation dataset by combining it with the UAV's positioning data). In this step, the detection time for a single battery group on the UAV is ≤10 minutes.

[0035] Step 3: Multi-source data fusion analysis

[0036] The data fusion and analysis module receives the infrared thermal imaging data and the spatiotemporally correlated multi-source data, preprocesses the infrared image (using adaptive noise reduction and temperature calibration algorithms), locates abnormal areas using a thermal anomaly identification algorithm, determines the anomaly type based on electrical parameters, and calculates the Battery Health Index (BHI). The anomaly type is determined as follows: when heating is detected at the wiring terminals, it is determined to be an increase in contact resistance; when overheating of a single battery cell is detected, it is determined to be an abnormal internal resistance. The accuracy rate of thermal anomaly identification in this step is ≥95%.

[0037] Step 4: Anomaly Warning and Report Generation

[0038] The data fusion analysis module triggers tiered early warnings based on BHI values ​​and / or local temperature and rate of temperature rise. Tiered early warnings include: Level 1 warning: BHI < 60 or local temperature ≥ 60℃, warning information will be pushed to ship maintenance personnel in real time; Level 2 warning: 60≤BHI<80 or temperature rise rate≥0.3℃ / charge-discharge cycle, generate a warning reminder.

[0039] Simultaneously, a standardized test report is automatically generated. This report includes an anomaly location diagram, a temperature distribution heatmap, an electrical parameter comparison table, a Battery Health Index (BHI), and handling recommendations, and can be exported in PDF or Word format.

[0040] Step 5: Data Archiving and Trend Tracking

[0041] The ground terminal uploads the detection data to the cloud database, and the cloud health management module generates a summary report on battery health status regularly (e.g., monthly), including temperature change curves, health index decay trend charts, and long-term warning records.

[0042] Based on the above, the present invention can obtain the following technical advantages: (1) In response to the special scenario of the closed and narrow battery compartment of ships, this invention optimizes the flight control and obstacle avoidance mechanism of UAVs, solves the pain point that traditional UAVs cannot enter the compartment for operation, realizes non-contact full coverage detection, thereby shortening the detection time of a single battery to the minute level (≤10 minutes), which is more than 80% more efficient than manual detection, and completely eliminates the safety risks of personnel entering the narrow battery compartment (such as electric shock, chemical corrosion, etc.).

[0043] (2) This invention establishes a precise correlation between infrared temperature data and battery electrical parameters and spatial location, breaking through the limitation that single infrared detection cannot distinguish between "normal temperature rise" and "fault heating", thus improving the accuracy of thermal anomaly identification to over 95%, and can detect early slight thermal anomalies (temperature rise of 3-5℃) 3-5 charge-discharge cycles in advance.

[0044] (3) Based on the ship battery fault feature library (including 6 typical faults) and the three-dimensional health assessment model (temperature + electrical + life), this invention realizes intelligent classification of thermal anomalies and quantification of health status, and provides targeted handling suggestions to facilitate quick decision-making by operation and maintenance personnel.

[0045] (4) This invention realizes the systematic management of detection data, supports full life cycle tracking and trend prediction, can extend battery life by 10%-15%, and reduce the risk of ship downtime due to battery failure.

[0046] (5) This invention is technologically mature and highly practical, and can be widely applied to various ship scenarios. For example, it is suitable for daily inspection and fault diagnosis of power lithium battery packs for container ships and bulk carriers; it provides rapid testing solutions for emergency battery packs of offshore wind power maintenance vessels and rescue vessels; it provides battery inspection tools for ship repair shops, improving maintenance efficiency and quality; and it provides non-invasive testing methods for maritime authorities, improving the coverage and efficiency of ship battery safety supervision. In addition, it can also be used for structural health inspections before ship insurance underwriting and accident defect inspections before claims settlement, providing quantitative and authoritative inspection data to provide a scientific basis for insurance underwriting and claims settlement.

[0047] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A ship battery inspection system based on UAV infrared thermal imaging, characterized in that, include: The drone is used to fly autonomously inside the ship's battery compartment. The drone is equipped with an infrared thermal imager and a visible light camera to collect infrared thermal imaging data and visible light images of the battery pack, and sends the infrared thermal imaging data to a data fusion analysis module. The battery parameter synchronous acquisition module is used to synchronously acquire the battery's electrical parameters through the battery management system interface or wireless sensor node, and to perform spatiotemporal correlation between the electrical parameters and the infrared thermal imaging data acquired by the UAV to obtain spatiotemporally correlated multi-source data, which is then sent to the data fusion analysis module. The data fusion and analysis module is used to receive the infrared thermal imaging data and the spatiotemporally correlated multi-source data, preprocess the infrared image, locate the abnormal area through the thermal anomaly identification algorithm, and calculate the battery health index (BHI) in combination with the electrical parameters to trigger a graded early warning. The cloud-based health management module is used to receive the test results output by the data fusion and analysis module, establish a ship battery test database, and realize data archiving, trend analysis, and multi-terminal access.

2. The system according to claim 1, characterized in that, The infrared thermal imager carried by the UAV has a temperature measurement range of -20℃ to 150℃, a temperature measurement accuracy of ±2%, and a temperature resolution of 0.1℃. The drone is also equipped with an ultrasonic obstacle avoidance sensor and a visual obstacle avoidance module, with a minimum safe distance of ≥100cm from the battery pack; The drone's detection time for a single battery pack is ≤10 minutes, and the data transmission delay is ≤200ms.

3. The system according to claim 1, characterized in that, The electrical parameters collected by the battery parameter synchronous acquisition module include the total voltage of the battery pack, the voltage of each individual cell, the charging and discharging current, the internal resistance, and the number of cycles, with an acquisition frequency ≥1Hz; The spatiotemporal correlation refers to: assigning a unified timestamp to infrared thermal imaging data and electrical parameters, with a timestamp synchronization error of ≤100ms, and establishing a three-dimensional correlation dataset containing temperature distribution, electrical parameters, and spatial location by combining UAV positioning data.

4. The system according to claim 1, characterized in that, The data fusion and analysis module also includes an environmental parameter adaptation unit, which is used to synchronously collect temperature and humidity data and ventilation status data in the battery compartment as the environmental correction basis for thermal anomaly judgment; the thermal anomaly identification algorithm is based on the ship battery thermal anomaly feature library, which includes six typical faults including local overheating, terminal heating and poor battery consistency.

5. The system according to claim 1, characterized in that, The data fusion and analysis module automatically marks the location of the anomaly, the highest temperature at the anomaly location, and the rate of temperature rise using the thermal anomaly identification algorithm, with a thermal anomaly identification accuracy of ≥95%; the battery health index (BHI) is calculated using the following formula: BHI = 0.4×Ts + 0.3×Ep + 0.3×Lc, Wherein, Ts is the temperature state coefficient, with a value of 0-100, calculated based on the degree of thermal anomaly and temperature consistency; Ep is the electrical parameter coefficient, with a value of 0-100, calculated based on voltage consistency and internal resistance change rate; Lc is the life coefficient, with a value of 0-100, calculated based on the number of cycles and the proportion of design life. The data fusion and analysis module classifies data according to the BHI value: BHI≥80 indicates a healthy state, 60≤BHI<80 indicates a state of concern, and BHI<60 indicates a state of warning.

6. The system according to claim 1, characterized in that, The cloud-based health management module establishes a ship battery testing database with a data retention period of ≥5 years, and supports multi-dimensional retrieval by ship name, battery pack number, or testing time. The cloud-based health management module automatically generates a temperature change curve and a health index decay trend chart for a single battery pack, and triggers an early warning when the temperature rise rate is ≥0.5℃ / charge-discharge cycle. The cloud-based health management module supports login via computer, tablet, and mobile phone, and also supports remotely issuing detection tasks.

7. A method for detecting ship batteries based on UAV infrared thermal imaging, implemented using the system described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step 1: Import the structural diagram of the target ship's battery compartment or select a ship type template on the ground terminal, set the inspection area, waypoint density and safety distance, generate the inspection route, and configure the BMS interface parameters; Step 2: The UAV flies along the planned inspection route, collecting infrared thermal imaging data and visible light images in real time; the battery parameter synchronous acquisition module collects battery electrical parameters and environmental data synchronously through the BMS interface or wireless sensor node, and completes spatiotemporal correlation annotation; Step 3: The data fusion analysis module receives the infrared thermal imaging data and the spatiotemporally correlated multi-source data, preprocesses the infrared image, locates the abnormal area through the thermal anomaly identification algorithm, determines the anomaly type by combining electrical parameters, and calculates the battery health index (BHI). Step 4: The data fusion and analysis module triggers graded early warnings based on the BHI value and / or local temperature and temperature rise rate, and automatically generates a standardized test report; Step 5: The ground terminal uploads the detection data to the cloud database, and the cloud health management module regularly generates a summary report on the battery health status.

8. The method according to claim 7, characterized in that, The method for determining the type of abnormality in step 3 is as follows: when the data fusion analysis module detects that the wiring terminal is overheating, it is determined that the contact resistance has increased; when it detects that a single battery cell is overheating, it is determined that the internal resistance is abnormal. The tiered early warning system in step 4 includes: The Level 1 warning condition is BHI < 60 or local temperature ≥ 60℃. The data fusion and analysis module pushes the warning information to the ship maintenance personnel in real time. The conditions for a Level II warning are 60 ≤ BHI < 80 or a temperature rise rate ≥ 0.3℃ / charge-discharge cycle. The data fusion analysis module generates a reminder to pay attention.

9. The method according to claim 7, characterized in that, The standardized test report includes an abnormal location diagram, a temperature distribution heat map, an electrical parameter comparison table, a battery health index (BHI), and handling recommendations, and can be exported in PDF or Word format.

10. The method according to claim 7, characterized in that, The waypoint density in step 1 is 5-10 cm / point; The single-cell battery detection time for the UAV in step 2 is ≤10 minutes; The accuracy rate of thermal anomaly identification in step 3 is ≥95%.

Images