A method and system for early warning of faults in the operating state of a drone based on data analysis

Abstract

This invention relates to the field of UAV operational status analysis technology, and discloses a data analysis-based method and system for early warning of UAV operational status faults. The method includes: collecting dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the UAV during its movement; directly assessing operational status faults of the UAV based on the dynamic parameter data to obtain a first assessment result; conducting adversarial assessment of operational status faults of the UAV based on the path control data, actual trajectory, and real-time energy consumption to obtain a second assessment result; and summarizing and analyzing the first and second assessment results to generate a fault warning strategy for the UAV's operational status. This significantly reduces the probability of misjudgment and missed judgment, enabling the UAV to have higher safety and reliability in complex tasks, and improving operational efficiency and task completion quality.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) operational status analysis technology, specifically to a method and system for early warning of UAV operational status faults based on data analysis. Background Technology

[0002] With the rapid development of drone technology, drones have been widely used in logistics, inspection, agricultural monitoring, and environmental protection. However, drones often face multiple challenges during operation due to complex external environments (such as wind, humidity, and temperature) and internal systems (such as batteries, power systems, and flight control systems). The uncertainty of their operational status and potential failure risks have become significant issues restricting their widespread application. Abnormalities in drone operation can lead to mission failures or even safety accidents. Therefore, accurately monitoring drone operational status and promptly detecting and warning of potential faults has become a core research direction in the field of drone technology.

[0003] Existing technologies for fault assessment of UAV operational status often employ single data source analysis, such as anomaly detection based on attitude or energy consumption data. However, these methods fail to comprehensively integrate multidimensional data and neglect the interaction of various factors under complex operating conditions, resulting in low fault identification and accuracy rates. Furthermore, traditional assessment methods are insufficient in distinguishing between external environmental disturbances (such as strong winds) and UAV system malfunctions, easily leading to misjudgments or missed detections. Therefore, there is an urgent need for a UAV fault early warning method based on multidimensional data fusion analysis, combining direct assessment and adversarial assessment to achieve accurate fault identification and rapid early warning. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing methods for early warning of drone operational status faults cannot clearly distinguish whether the fault is caused by the drone itself or by wind; and they cannot analyze different control methods separately.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for early warning of unmanned aerial vehicle (UAV) operational status faults based on data analysis, comprising:

[0007] Collect dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the drone during its movement;

[0008] Based on the dynamic parameter data, a direct assessment of the UAV's operational status faults is performed to obtain a first assessment result;

[0009] Based on the path control data, actual trajectory, and real-time energy consumption, an anti-fault assessment of the UAV's operational status is conducted to obtain a second assessment result.

[0010] The first evaluation result and the second evaluation result are summarized and analyzed to generate a fault early warning strategy for the drone's operating status.

[0011] As a preferred embodiment of the data analysis-based UAV operational status fault early warning method of the present invention, the dynamic parameter data includes attitude data and power system data;

[0012] The attitude data includes attitude angle data, angular velocity data, acceleration data, and vibration frequency;

[0013] The power system data includes the speed of each motor, the temperature of each motor, and the actual power of each motor.

[0014] The path control data includes control data for the UAV's flight trajectory. If the UAV is flying along a predetermined route, the path control data indicates control over the predetermined route. If the UAV is under real-time control, the path control data indicates that the UAV has received control commands from real-time control.

[0015] The actual trajectory includes real-time updates of the drone's trajectory through real-time positioning.

[0016] As a preferred embodiment of the data analysis-based UAV operation status fault early warning method of the present invention, the direct assessment includes: for each dynamic parameter data, a preset fluctuation range is set, and the real-time dynamic parameter data is compared with the normal fluctuation range corresponding to each data type. If it is not within the range, the UAV operation status is judged to be abnormal; otherwise, the UAV operation status is not judged.

[0017] The judgment that the drone's operating status is abnormal is taken as the first evaluation result.

[0018] As a preferred embodiment of the data analysis-based UAV operational status fault early warning method of the present invention, the countermeasure evaluation includes countermeasures against the actual trajectory and real-time energy consumption based on the path control data, thereby enabling the UAV to verify operational status faults.

[0019] The verification results will be used as the second evaluation result;

[0020] The confrontation process is divided into: trajectory confrontation analysis and energy consumption confrontation analysis;

[0021] The objective of the trajectory adversarial analysis is to verify whether the drone's flight trajectory conforms to the predetermined objective or real-time control commands.

[0022] The goal of the energy consumption countermeasure analysis is to verify whether the current energy consumption of the UAV matches its flight mission and environmental conditions.

[0023] As a preferred embodiment of the data analysis-based UAV operational status fault early warning method described in this invention, the trajectory countermeasure analysis includes: Case 1: Flying along a predetermined route, analyzing the real-time shortest distance between the UAV and the predetermined route through real-time positioning. When the drone deviates from the predetermined route, Record and Arranged in time sequence, the recording ends when the drone returns to the predetermined route and travels a stable preset distance S; and the recording from the beginning of deviation from the predetermined route to the return to the predetermined route and the stable travel of the preset distance S is output. sequence When the drone deviates from the predetermined route again, the sequence will be recorded and generated again.

[0024] in, This represents the sequence of records of the shortest distances when the nth deviation from the predetermined route is made;

[0025] Analysis is performed using pre-trained generator 1 and discriminator 1 from the adversarial network. Input: and Wind parameters in the corresponding time series Output: The probability of drone malfunction causing offset in the corresponding time series;

[0026] in, express The corresponding maximum wind speed in the time series. express The minimum wind speed in the corresponding time series. express The corresponding average wind speed in the time series;

[0027] Scenario 2: Flight is conducted using real-time control. The degree of control implementation is calculated by analyzing each control signal. Specifically, this involves using a pre-trained generator 2 and discriminator 2 from an adversarial network for analysis. Input: and Wind speed at the moment of occurrence; Output: In Predicting the displacement vector of the UAV under control ;

[0028] exist In the sequence, arbitrarily select the displacement vectors corresponding to L consecutive control signals, add the L displacement vectors together to obtain L consecutive control signals, and predict the achievable displacement control result. Simultaneously, acquire L consecutive control signals and their position information at the start and end points, and construct the actual displacement based on the position information. ;

[0029] The degree to which the control behavior is implemented is expressed as follows:

[0030]

[0031] In scenario two, output the maximum value of all sampled results. The average of all sampled results ;

[0032] in, This represents the m-th control signal; L represents the fluctuation value, which is any integer between the preset maximum and the preset minimum value. This indicates the degree of control behavior implementation corresponding to L consecutive control signals;

[0033] The energy consumption adversarial analysis includes analysis using a pre-trained generator 3 and discriminator 3 in an adversarial network. Inputs include: the actual trajectory acquired in real-time by the UAV, and wind parameters from the start of its movement to the current moment. Output: The probability of an abnormal excessive energy consumption.

[0034] in, This represents the maximum wind force from the start of the motion to the current moment. This represents the minimum wind force from the start of the motion to the current moment. This represents the average wind force from the start of the movement to the current moment.

[0035] As a preferred embodiment of the data analysis-based UAV operation status fault early warning method of the present invention, the summary analysis includes obtaining the first evaluation result and the second evaluation result respectively under the two types of flight states: the predetermined route and the real-time control.

[0036] The obtained evaluation results are compared with the corresponding preset thresholds to obtain the judgment result for each output element in the first evaluation result and the second evaluation result;

[0037] Based on the output judgment result, the fault type is matched;

[0038] Fault labels are obtained from the assessment of the drone's operational status.

[0039] As a preferred embodiment of the data analysis-based UAV operation status fault early warning method of the present invention, the fault early warning strategy includes setting a corresponding early warning strategy for each fault label, issuing the control command in the early warning strategy as the first priority to the UAV, and executing it.

[0040] A data analysis-based UAV operational status fault early warning system employing any of the methods described in this invention, characterized in that:

[0041] The data acquisition unit collects dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the UAV during its movement.

[0042] The evaluation unit directly evaluates the operational status faults of the UAV based on the dynamic parameter data, and obtains a first evaluation result.

[0043] The analysis unit performs an operational status fault assessment on the UAV based on the path control data, actual trajectory, and real-time energy consumption, and obtains a second assessment result.

[0044] The early warning unit summarizes and analyzes the first evaluation result and the second evaluation result to generate a fault early warning strategy for the UAV's operating status.

[0045] A computer device includes: a memory and a processor; the memory stores a computer program, wherein: when the processor executes the computer program, it implements the steps of the method described in any one of the present invention.

[0046] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of the present invention.

[0047] The beneficial effects of this invention are as follows: The data analysis-based UAV operational status fault early warning method provided by this invention achieves accurate fault detection and real-time early warning of UAV operational status by combining direct evaluation of dynamic parameter data with adversarial evaluation of path and energy consumption. Utilizing multi-source data fusion and adversarial network technology improves the ability to distinguish between external interference and internal faults in complex environments, significantly reducing the probability of false positives and false negatives. Simultaneously, combining fault tags to generate personalized early warning strategies enhances the safety and reliability of UAVs in complex tasks, improving operational efficiency and mission completion quality. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 The first embodiment of the present invention provides an overall flowchart of a method for early warning of unmanned aerial vehicle (UAV) operational status faults based on data analysis. Detailed Implementation

[0050] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0051] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for early warning of unmanned aerial vehicle (UAV) operational status faults based on data analysis is provided, comprising:

[0052] S1: Collects dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the drone during its movement.

[0053] Furthermore, the dynamic parameter data includes attitude data and power system data; the attitude data includes attitude angle data, angular velocity data, acceleration data, and vibration frequency; the power system data includes the rotational speed of each motor, the temperature of each motor, and the actual power of each motor. The path control data includes control data for the UAV's flight trajectory. If the UAV is flying along a predetermined route, the path control data indicates control of the predetermined route; if the UAV is under real-time control, the path control data indicates that the UAV has received control commands from real-time control. The actual trajectory includes real-time updates to the UAV's trajectory through real-time positioning.

[0054] By collecting attitude and power system data, the operational attitude (such as flight angle and rotation speed) and power performance (such as motor speed, temperature, and power) of the UAV can be monitored in real time, providing core evidence for judging the stability and health status of the UAV. Combining path control data with the actual trajectory of real-time positioning enables a comprehensive understanding of whether the UAV is executing its flight mission as planned and identifies potential trajectory deviations or path loss of control issues.

[0055] By combining a predetermined route or real-time control commands, the system differentiates between trajectory deviations caused by external environmental interference (such as wind or obstacles) and problems caused by internal system malfunctions (such as motor failure or control lag). Through correlation analysis between real-time energy consumption data and trajectory deviations, it determines whether high energy consumption is caused by external factors (such as strong winds) or system anomalies (such as battery or motor failures). By real-time acquisition and updating of dynamic parameters and trajectory data, the system provides real-time monitoring capabilities, enabling rapid detection of abnormal changes during flight. It provides real-time input data, allowing subsequent early warning algorithms to respond quickly under dynamic conditions, ensuring the UAV's flight safety. The various types of data collected in real-time can not only be used for fault early warning but also provide a basis for optimizing the UAV's flight control algorithms. For example, by analyzing changes in attitude data and power system data, flight control parameters can be improved, enhancing flight performance and stability.

[0056] S2: Based on the dynamic parameter data, a direct assessment of the UAV's operational status faults is performed to obtain a first assessment result.

[0057] For each dynamic parameter data, a preset fluctuation range is defined. The real-time dynamic parameter data is compared with the normal fluctuation range corresponding to its respective data type. If the data is outside the range, the drone's operating status is determined to be abnormal; otherwise, no judgment is made on the drone's operating status. The determination of the drone's abnormal operating status is output as the first evaluation result.

[0058] It's important to note that by real-time monitoring of UAV dynamic parameter data and comparing it with preset normal fluctuation ranges, the system can quickly identify and determine whether the UAV's operating status is abnormal, thus achieving a preliminary assessment of potential faults. By collecting dynamic parameters (such as attitude angles, angular velocity, and motor speeds) in real time and comparing them with normal ranges, it can accurately capture fluctuations exceeding the normal range during operation, effectively detecting whether the UAV has attitude abnormalities, power system abnormalities, or other operational anomalies. This step uses static thresholds or dynamically adjusted normal fluctuation ranges as benchmarks to ensure the scientific validity and operability of the assessment standards, while avoiding misjudgments due to slight interference from the external environment or occasional noise, improving the accuracy and stability of the assessment results. The first assessment result generated through direct evaluation can quickly filter out obvious abnormal operating states, providing input for subsequent, more complex adversarial assessments, shortening the response time for fault identification, and ensuring the comprehensiveness and efficiency of the system's early warning.

[0059] S3: Based on the path control data, actual trajectory, and real-time energy consumption, conduct an anti-fault assessment of the UAV's operational status to obtain a second assessment result.

[0060] It should be noted that the adversarial assessment includes, based on the path control data, performing adversarial analysis on the actual trajectory and real-time energy consumption respectively, to verify the operational status faults of the UAV. The verification result is then used as the second assessment result. The adversarial process is divided into: trajectory adversarial analysis and energy consumption adversarial analysis.

[0061] Furthermore, the objective of the trajectory countermeasure analysis is to verify whether the UAV's flight trajectory conforms to the predetermined objective or real-time control commands. The objective of the energy consumption countermeasure analysis is to verify whether the UAV's current energy consumption matches its flight mission and environmental conditions.

[0062] Specifically, the trajectory adversarial analysis includes, in case one: flying along a predetermined route, analyzing the real-time shortest distance between the UAV and the predetermined route through real-time positioning. When the drone deviates from the predetermined route ( When the value exceeds the preset value, it is considered a deviation. Record and Arranged in time sequence, the recording ends when the drone returns to the predetermined route and travels a stable preset distance S; and the recording from the beginning of deviation from the predetermined route to the return to the predetermined route and the stable travel of the preset distance S is output. sequence When the drone deviates from the predetermined route again, the sequence is recorded and generated again.

[0063] in, This represents the sequence of records of the shortest distances when the nth deviation from the predetermined route occurs.

[0064] Analysis is performed using pre-trained generator 1 and discriminator 1 from the adversarial network. Input: and Wind parameters in the corresponding time series Output: The probability of drone malfunction causing offset in the corresponding time series.

[0065] in, express The corresponding maximum wind speed in the time series. express The minimum wind speed in the corresponding time series. express The average wind speed in the corresponding time series.

[0066] The key point is that by analyzing the trajectory deviation of a drone during its flight along a predetermined route, and considering the influence of external environmental factors such as wind, an adversarial network model is used to accurately assess the probability of the drone deviating from its predetermined route due to its own malfunction, thus distinguishing between deviations caused by external interference and internal faults. Using wind parameters (such as maximum, minimum, and average wind speeds) as input features, combined with the trajectory deviation sequence, the adversarial network model is trained to effectively model the degree of influence of the external environment (wind) on trajectory deviation. Based on this, through adversarial learning between the generator and discriminator, the potential fault risks of the drone system itself during the deviation process are assessed, and the probability of deviation caused by a fault is output. This design avoids the simple reliance on trajectory deviation judgment in traditional methods, and can accurately identify whether the deviation is due to normal external environmental factors (such as strong winds) or abnormal deviation due to internal system factors (such as motor failure or abnormal control response).

[0067] Scenario 2: Flight is conducted using real-time control. The degree of control implementation is calculated by analyzing each control signal. Specifically, this involves using a pre-trained generator 2 and discriminator 2 from an adversarial network for analysis. Input: and Wind speed at the moment of occurrence; Output: In Predicting the displacement vector of the UAV under control .

[0068] exist In the sequence, arbitrarily select the displacement vectors corresponding to L consecutive control signals, add the L displacement vectors together to obtain L consecutive control signals, and predict the achievable displacement control result. Simultaneously, acquire L consecutive control signals and their position information at the start and end points, and construct the actual displacement based on the position information. .

[0069] The degree to which the control behavior is implemented is expressed as follows:

[0070]

[0071] In scenario two, output the maximum value of all sampled results. The average of all sampled results .

[0072] in, This represents the m-th control signal; L represents the fluctuation value, which is any integer between the preset maximum and the preset minimum value. This indicates the degree to which the control behavior is achieved corresponding to L consecutive control signals.

[0073] By inputting each control signal and its corresponding wind speed, an adversarial network is used to generate the theoretical displacement vector of the UAV under the action of that control signal, and this vector is compared with the actual displacement. This process effectively distinguishes the difference between the theoretical effect and the actual execution effect of the control signal, helping to identify control behavior deviations caused by internal faults (such as flight control system delays or motor failures). By arbitrarily selecting the displacement vectors corresponding to L consecutive control signals, the predicted result of the control signal over a period of time is obtained through accumulation calculation, and compared with the actual displacement, thus comprehensively evaluating the degree of control behavior realization within a specific time period. This design overcomes the limitation of a single signal potentially being affected by accidental interference, and more comprehensively reflects the overall performance of the control system. By combining the input of real-time wind parameters (such as wind speed), the correlation analysis between control behavior realization deviations and the influence of external environmental factors can be performed, distinguishing between deviations caused by external interference such as strong winds and deviations caused by internal control system faults. This modeling method reduces false positives and false negatives, improving the accuracy of evaluating control behavior in complex flight environments.

[0074] By outputting the maximum and average values ​​of all sampled results, the fluctuation range and stability of the control behavior are quantified. This provides data support for analyzing the overall performance of the control system (such as response delay and insufficient accuracy), and also provides optimization directions for subsequent improvements to flight control algorithms or hardware configurations.

[0075] The energy consumption adversarial analysis includes analysis using a pre-trained generator 3 and discriminator 3 in an adversarial network. Inputs include: the actual trajectory acquired in real-time by the UAV, and wind parameters from the start of its movement to the current moment. Output: The probability of an abnormal excessive energy consumption.

[0076] in, This represents the maximum wind force from the start of the motion to the current moment. This represents the minimum wind force from the start of the motion to the current moment. This represents the average wind force from the start of the movement to the current moment.

[0077] S4: Summarize and analyze the first evaluation results and the second evaluation results to generate a fault warning strategy for the UAV's operating status.

[0078] Under both the predetermined route and real-time control flight conditions, the first evaluation result and the second evaluation result are acquired. The acquired evaluation results are compared with corresponding preset thresholds to obtain a judgment result for each output element in the first and second evaluation results. Based on the output judgment results, the fault type is matched to obtain a fault label for the UAV's operational status assessment.

[0079] Specifically, in this embodiment, under the predetermined route (under this condition, it is not necessary to output the probability of deviation due to excessive wind): When the first evaluation result shows no abnormality, if the probability of deviation caused by UAV malfunction is less than the corresponding threshold, and the probability of abnormal excessive energy consumption is less than the corresponding threshold, then it is determined that there is no abnormality. When the first evaluation result shows no abnormality, if the probability of deviation caused by UAV malfunction is greater than the corresponding threshold, and the probability of abnormal excessive energy consumption is less than the corresponding threshold, then it is determined that the system is abnormal. When the first evaluation result shows no abnormality, if the probability of deviation caused by UAV malfunction is less than the corresponding threshold, and the probability of abnormal excessive energy consumption is greater than the corresponding threshold, then it is determined that the power system is abnormal. When the first evaluation result shows no abnormality, if the probability of deviation caused by UAV malfunction is greater than the corresponding threshold, and the probability of abnormal excessive energy consumption is greater than the corresponding threshold, then it is determined that there is a mixed abnormality. When the first evaluation result shows abnormality, if the probability of abnormal excessive energy consumption is less than the corresponding threshold, then it is determined that the power system is abnormal; if the probability of abnormal excessive energy consumption is greater than the corresponding threshold, then it is determined that there is a mixed abnormality.

[0080] During flight under real-time control: When the first evaluation result shows no abnormality, if the outputs of Case Two are all less than the corresponding threshold, and the probability of excessive energy consumption is less than the corresponding threshold, then no abnormality is determined. When the first evaluation result shows no abnormality, if any value of the output of Case Two is greater than the corresponding threshold, and the probability of excessive energy consumption is less than the corresponding threshold, then a system abnormality is determined. When the first evaluation result shows no abnormality, if any value of the output of Case Two is greater than the corresponding threshold, and the probability of excessive energy consumption is greater than the corresponding threshold, then a mixed abnormality is determined. When the first evaluation result shows no abnormality, if the outputs of Case Two are all less than the corresponding threshold, and the probability of excessive energy consumption is greater than the corresponding threshold, then the power system is abnormal. When the first evaluation result shows an abnormality, if the probability of excessive energy consumption is greater than the corresponding threshold, then a mixed abnormality is determined; if the probability of excessive energy consumption is less than the corresponding threshold, then the power system is determined to be abnormal.

[0081] When only the maximum value of all sampled results is output. The value is greater than the corresponding threshold, while the average value of all samples is greater than the threshold. If the value is less than the corresponding threshold, a sudden anomaly is determined.

[0082] The fault warning strategy includes setting a corresponding warning strategy for each fault label, issuing the control command in the warning strategy as the first priority to the UAV, and executing it.

[0083] In one optional embodiment, the flight is immediately grounded when a mixed anomaly occurs; when a sudden anomaly occurs, it is temporarily ignored and recorded; when a system anomaly or a power system anomaly occurs, the flight is grounded after the current task is completed; all of these can be modified according to preset plans.

[0084] It's worth noting that by combining direct and adversarial assessment results, abnormal UAV operational statuses are categorized into different types (no abnormality, system abnormality, power system abnormality, mixed abnormality, and sudden abnormality), with separate judgment logic and conditions set according to flight mode. This refined classification design avoids misjudgments or omissions that may occur with traditional single assessment methods, ensuring accurate identification of problem sources (such as wind interference, power system failure, or control system abnormality) in complex operating environments. Based on the abnormality type, multi-level response strategies are provided (such as immediate grounding, grounding after mission completion, and continued recording and observation), enabling graded handling of abnormalities. In particular, the high-priority response strategy (immediate grounding) for mixed abnormalities (coexistence of internal system and power system problems) and the low-priority strategy (temporarily ignoring and recording) for sudden abnormalities ensure flight safety while avoiding excessive intervention in flight missions due to non-critical issues.

[0085] By linking fault tags with early warning strategies, the system can quickly generate and execute corresponding control commands after fault diagnosis, ensuring the operational stability of the UAV in complex missions. Without affecting mission completion, it can reasonably handle system or power system anomalies (such as grounding the UAV after mission completion), improving mission execution efficiency.

[0086] On the other hand, this embodiment also provides a data analysis-based UAV operational status fault early warning system, which includes:

[0087] The data acquisition unit collects dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the UAV during its movement.

[0088] The evaluation unit directly evaluates the operational status faults of the UAV based on the dynamic parameter data, and obtains a first evaluation result.

[0089] The analysis unit performs an operational status fault assessment on the UAV based on the path control data, actual trajectory, and real-time energy consumption, and obtains a second assessment result.

[0090] The early warning unit summarizes and analyzes the first evaluation result and the second evaluation result to generate a fault early warning strategy for the UAV's operating status.

[0091] If the above functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0092] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0093] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0094] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0095] Example 2 is an embodiment of the present invention, which provides a method for early warning of unmanned aerial vehicle (UAV) operational status faults based on data analysis. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0096] The test subjects were two identical UAVs: UAV A used the fault warning method of this invention, while UAV B used the traditional single dynamic parameter analysis method for comparison. The test environment was set in a complex flight mission scenario, including both predetermined route flight and real-time control modes. Specific flight mission conditions included a flight altitude of 50 meters, wind speed variation range of 3-10 m / s, a total flight distance of 5 kilometers, and a payload of 1 kg. During the experiment, the following data were collected in real time: attitude data (pitch angle, roll angle), power system data (motor speed, power consumption), path deviation data (shortest deviation distance, deviation time), and real-time energy consumption.

[0097] Before the experiment began, the flight control parameters of each UAV were set to factory default values. For Type A UAVs, the fault early warning method described in the invention was loaded, including modules for direct evaluation of dynamic parameters, trajectory countermeasure analysis, and energy consumption countermeasure analysis. Type B UAVs only used traditional methods, judging faults by monitoring fixed thresholds of dynamic parameters.

[0098] The experiment consists of the following steps:

[0099] Predetermined route flight mission: The UAV flies along a preset route, collecting dynamic parameter data and trajectory deviation data in real time. When the deviation distance exceeds 1 meter or the energy consumption exceeds 20% of the expected value, the deviation time and the probability of energy consumption anomalies are recorded.

[0100] Real-time control task: The UAV navigates to random target points via ground control commands, and monitors the execution of control signals and the degree of displacement achievement in real time. When the degree of control signal achievement is below 80%, the frequency and impact range of its occurrence are recorded.

[0101] Data recording and analysis: After the experiment, the operation data of the two drones were collected, and their fault warning success rate, false judgment rate and impact on task completion efficiency were statistically analyzed. The data is recorded in Table 1.

[0102] Table 1 Data Record Table

[0103]

[0104] The experimental data shows that the fault early warning method of this invention exhibits significant advantages in fault identification accuracy, flight stability, and mission completion efficiency. Through analysis and comparison, the following conclusions can be drawn:

[0105] The fault warning success rate of the Type A UAV reached 98%, far exceeding the 82% of the Type B UAV. This result demonstrates that the present invention, by combining direct evaluation of dynamic parameters with adversarial evaluation of path and energy consumption, can effectively capture potential faults. Furthermore, the false positive rate of the Type A UAV was only 2%, significantly lower than the 10% of the Type B UAV, showing that the present invention has a stronger ability to distinguish between external interference and internal faults in complex environments.

[0106] The average deviation distance of the Type A UAV is 0.6 meters, significantly lower than the 1.5 meters of the Type B UAV; the deviation time is also reduced from 28 seconds using the traditional method to 12 seconds. This indicates that the trajectory countermeasure analysis module of the present invention can identify and correct trajectory deviations in a timely manner, ensuring flight accuracy.

[0107] In terms of energy consumption, the average power consumption of the Type A drone is 280W, while that of the Type B drone is 310W. The energy consumption countermeasure analysis module of this invention reduces unnecessary power loss by adjusting the power system output in real time, while effectively avoiding abnormal situations of excessive energy consumption.

[0108] In real-time control mode, the control signal implementation rate of UAV A is 94%, higher than that of UAV B (85%). This indicates that the method of the present invention can accurately predict displacement response through dynamic analysis of control signals, thereby optimizing the execution effect of the control system.

[0109] The Type A UAV completed all test missions, while the Type B UAV achieved only 85% mission completion due to multiple malfunctions and deviations. This demonstrates that the present invention not only improves operational safety but also significantly enhances mission execution reliability.

[0110] The experimental results clearly demonstrate that the fault early warning method of this invention overcomes the limitations of traditional methods that rely on single data source analysis through multi-source data fusion analysis and intelligent evaluation, exhibiting significant innovation and advantages. Especially in complex flight missions, this invention can more effectively identify and warn of potential faults, ensuring the safety and mission completion rate of UAVs, and providing crucial support for the practical application of UAV technology.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for early warning of operational status faults in unmanned aerial vehicles (UAVs) based on data analysis, characterized in that, include: Collect dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the drone during its movement; Based on the dynamic parameter data, a direct assessment of the UAV's operational status faults is performed to obtain a first assessment result; Based on the path control data, actual trajectory, and real-time energy consumption, an anti-fault assessment of the UAV's operational status is conducted to obtain a second assessment result. The first evaluation result and the second evaluation result are summarized and analyzed to generate a fault early warning strategy for the UAV's operating status; The dynamic parameter data includes attitude data and power system data; The attitude data includes attitude angle data, angular velocity data, acceleration data, and vibration frequency; The power system data includes the speed of each motor, the temperature of each motor, and the actual power of each motor. The path control data includes control data for the UAV's flight trajectory. If the UAV is flying along a predetermined route, the path control data indicates control over the predetermined route. If the UAV is under real-time control, the path control data indicates that the UAV has received control commands from real-time control. The actual trajectory includes real-time updates of the drone's trajectory through real-time positioning; The direct assessment includes comparing the real-time dynamic parameter data with the normal fluctuation range corresponding to each data type within the preset fluctuation range of each dynamic parameter data. If the data is not within the range, the drone's operating status is judged to be abnormal. Conversely, the operating status of the drone is not judged; The judgment that the drone's operating status is abnormal is taken as the first evaluation result; The adversarial assessment includes, based on the path control data, performing adversarial assessments on the actual trajectory and real-time energy consumption respectively, to verify the operational status faults of the UAV; The verification results will be used as the second evaluation result; The confrontation process is divided into: trajectory confrontation analysis and energy consumption confrontation analysis; The objective of the trajectory adversarial analysis is to verify whether the drone's flight trajectory conforms to the predetermined objective or real-time control commands. The goal of the energy consumption countermeasure analysis is to verify whether the current energy consumption of the UAV matches its flight mission and environmental conditions.

2. The method for early warning of UAV operational status faults based on data analysis as described in claim 1, characterized in that: The trajectory adversarial analysis includes, in case one: flying along a predetermined route, analyzing the real-time shortest distance between the UAV and the predetermined route using real-time positioning. ; When the drone deviates from the predetermined route, Record and Arranged in time sequence, the recording ends when the drone returns to the predetermined route and travels a stable preset distance S; and the recording from the beginning of deviation from the predetermined route to the return to the predetermined route and the stable travel of the preset distance S is output. sequence When the drone deviates from the predetermined route again, the sequence will be recorded and generated again. in, This represents the sequence of records of the shortest distances when the nth deviation from the predetermined route is made; Analysis is performed using pre-trained generator 1 and discriminator 1 from the adversarial network. Input: and Wind parameters in the corresponding time series Output: The probability of drone malfunction causing offset in the corresponding time series; in, express The corresponding maximum wind speed in the time series. express The minimum wind speed in the corresponding time series. express The corresponding average wind speed in the time series; Scenario 2: Flight is conducted using real-time control. The degree of control implementation is calculated by analyzing each control signal. Specifically, this involves using a pre-trained generator 2 and discriminator 2 from an adversarial network for analysis. Input: and Wind speed at the moment of occurrence; Output: In Predicting the displacement vector of the UAV under control ; exist In the sequence, arbitrarily select the displacement vectors corresponding to L consecutive control signals, add the L displacement vectors together to obtain L consecutive control signals, and predict the achievable displacement control result. Simultaneously, acquire L consecutive control signals and their position information at the start and end points, and construct the actual displacement based on the position information. ; The degree to which the control behavior is implemented is expressed as follows: In scenario two, output the maximum value of all sampled results. The average of all sampled results ; in, This represents the m-th control signal; L represents the fluctuation value, which is any integer between the preset maximum and the preset minimum value. This indicates the degree of control behavior implementation corresponding to L consecutive control signals; The energy consumption adversarial analysis includes analysis using a pre-trained generator 3 and discriminator 3 in an adversarial network. Inputs include: the actual trajectory acquired in real-time by the UAV, and wind parameters from the start of its movement to the current moment. Output: The probability of an abnormal excessive energy consumption. in, This represents the maximum wind force from the start of the motion to the current moment. This represents the minimum wind force from the start of the motion to the current moment. This represents the average wind force from the start of the movement to the current moment.

3. The method for early warning of UAV operational status faults based on data analysis as described in claim 2, characterized in that: The summary analysis includes obtaining the first evaluation result and the second evaluation result under the two types of flight states: the predetermined route and the real-time control. The obtained evaluation results are compared with the corresponding preset thresholds to obtain the judgment result for each output element in the first evaluation result and the second evaluation result; Based on the output judgment result, the fault type is matched; Fault labels are obtained from the assessment of the drone's operational status.

4. The method for early warning of UAV operational status faults based on data analysis as described in claim 3, characterized in that: The fault warning strategy includes setting a corresponding warning strategy for each fault label, issuing the control command in the warning strategy as the first priority to the UAV, and executing it.

5. A data analysis-based early warning system for unmanned aerial vehicle (UAV) operational status, employing the method described in any one of claims 1-4, characterized in that: The data acquisition unit collects dynamic parameter data, path control data, actual trajectory, and real-time energy consumption of the UAV during its movement. The evaluation unit directly evaluates the operational status faults of the UAV based on the dynamic parameter data, and obtains a first evaluation result. The analysis unit performs an operational status fault assessment on the UAV based on the path control data, actual trajectory, and real-time energy consumption, and obtains a second assessment result. The early warning unit summarizes and analyzes the first evaluation result and the second evaluation result to generate a fault early warning strategy for the UAV's operating status.

6. A computer device, comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-4.

Images