Ota upgrade method and system based on unmanned aerial vehicle fault detection

Abstract

The application discloses an unmanned aerial vehicle fault detection-based OTA upgrading method and system, relates to the unmanned aerial vehicle control technical field, and discloses the unmanned aerial vehicle fault detection-based OTA upgrading method and system, real-time acquisition of unmanned aerial vehicle operation data is carried out, when the motor current overcurrent times in the unmanned aerial vehicle operation data are greater than the preset first number of times in a single flight process, preliminary fault data is generated; the auxiliary sensor data of the unmanned aerial vehicle is collected in real time through the airborne auxiliary sensor, the flight scene data of the unmanned aerial vehicle is acquired through the airborne GPS module, and the auxiliary sensor data, the flight scene data and the preliminary fault data are combined and analyzed to generate fault root source probability data, the software compatibility fault is accurately identified through multi-dimensional data fusion analysis, the upgrading operation is avoided from being mistakenly triggered, and the unmanned aerial vehicle operation reliability is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle control, in particular to an OTA upgrade method and system based on unmanned aerial vehicle fault detection. BACKGROUND

[0002] The current unmanned aerial vehicle control system generally adopts a fault detection mechanism based on threshold determination, for example, triggering a firmware upgrade instruction when the number of overcurrent faults detected exceeds a preset threshold. However, this method has significant defects: the fault diagnosis dimension is single, and it cannot distinguish between software compatibility faults and abnormalities caused by hardware or environmental factors. Specifically, the prior art only relies on the cumulative number of overcurrent faults as the basis for upgrading, ignoring the correlation analysis of multi-source data (such as temperature fluctuations, voltage abnormalities, and flight scene characteristics), resulting in a high misjudgment rate. For example, transient overcurrent caused by motor overheating may be mistakenly attributed to software defects, triggering unnecessary upgrade processes. This not only wastes network and computing resources, but also affects job efficiency due to interruptions during the upgrade process. In addition, the traditional upgrade mechanism lacks dynamic adaptation ability to task scenarios, and cannot intelligently schedule upgrade timing according to the urgency of the job, further reducing the reliability of the system. The prior art also has the following problems: the upgrade decision-making process lacks a cloud collaboration mechanism and cannot effectively utilize historical fault databases for intelligent matching; the upgrade execution process lacks scene adaptability and cannot dynamically adjust the upgrade strategy according to the flight mode; the upgrade verification link is missing, and the stability of the system after upgrading cannot be ensured.

[0003] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide an OTA upgrade method and system based on unmanned aerial vehicle fault detection, aiming to avoid false triggering of upgrade operations and improve the reliability of unmanned aerial vehicle operations.

[0005] To achieve the above-mentioned purpose, the present application provides an OTA upgrade method based on unmanned aerial vehicle fault detection, which comprises:

[0006] Real-time acquisition of unmanned aerial vehicle operation data, generation of preliminary fault data when the number of motor current overflows in the unmanned aerial vehicle operation data during a single flight is greater than a preset first number;

[0007] Real-time acquisition of auxiliary sensor data of the unmanned aerial vehicle by an on-board auxiliary sensor, acquisition of flight scene data of the unmanned aerial vehicle by an on-board GPS module, and fusion analysis of the auxiliary sensor data, the flight scene data, and the preliminary fault data to generate fault root cause probability data; generation of a potential fault upgrade instruction when the software compatibility problem probability in the fault root cause probability data exceeds a preset probability;

[0008] sending the fault root probability data and the UAV model data to a cloud server for upgrade decision processing upon receiving the potential fault upgrade instruction or the user-triggered active upgrade instruction;

[0009] generating an upgrade permission instruction and obtaining real-time control mode data and throttle signal data based on the upgrade permission instruction to generate task urgency identification data upon receiving the upgrade permission response data returned by the cloud server;

[0010] controlling the UAV to return to the ground and perform an OTA upgrade operation when the task urgency identification data indicates that the UAV is in a takeoff idle mode, and postponing the OTA upgrade operation until the end of the work task when the task urgency identification data indicates that the UAV is in a work mode.

[0011] In an embodiment, after the step of sending the fault root probability data and the UAV model data to a cloud server for upgrade decision processing upon receiving the potential fault upgrade instruction or the user-triggered active upgrade instruction, the method further comprises:

[0012] generating a cancellation upgrade instruction and generating fault warning data based on the cancellation upgrade instruction upon receiving the cancellation upgrade response data returned by the cloud, and uploading the fault warning data to the cloud server to perform a fault warning operation.

[0013] In an embodiment, before the step of sending the fault root probability data and the UAV model data to a cloud server for upgrade decision processing upon receiving the potential fault upgrade instruction or the user-triggered active upgrade instruction, the method further comprises:

[0014] extracting motor current time series data and control signal time series data from the UAV operation data;

[0015] inputting the motor current time series data and the control signal time series data into a pre-trained prediction model to output fault occurrence probability data;

[0016] generating the potential fault upgrade instruction when the fault occurrence probability data exceeds a warning threshold.

[0017] In an embodiment, the step of generating fault root probability data by fusing the auxiliary sensor data, the flight scene data, and the preliminary fault data comprises:

[0018] matching the preliminary fault data with a historical fault database to generate fault mode similarity data;

[0019] extracting temperature data and voltage data in the auxiliary sensor data, and generating hardware exception identification data when the temperature data continuously exceeds a preset temperature threshold and the voltage data fluctuates beyond a preset fluctuation threshold;

[0020] extracting height data and positioning data in the flight scene data, and generating scene compatibility identification data when it is determined that a current flight scene conforms to a preset software conflict scene based on the height data and the positioning data;

[0021] generating the fault root probability data from the fault mode similarity data, the hardware exception identification data, and the scene compatibility identification data through a preset rule.

[0022] In an embodiment, the method further comprises:

[0023] obtaining real-time weather severity data through a weather interface;

[0024] dynamically adjusting a preset probability of the software compatibility problem probability according to the weather severity data.

[0025] In an embodiment, based on the allowed upgrade instruction, real-time control mode data and throttle signal data are obtained, and the step of generating task urgency identification data comprises:

[0026] when the real-time control mode data is an automatic route mode, the throttle signal data exceeds a preset operation threshold and reaches a preset time length, generating an operation mode identification as the task urgency identification data;

[0027] when the control mode data is a manual mode and the throttle signal data is lower than a preset hovering threshold, generating an idle speed mode identification as the task urgency identification data.

[0028] In an embodiment, the upgrade determination process comprises:

[0029] the cloud server matches the fault root probability data with an incremental upgrade file library to generate file existence identification data;

[0030] when the file existence identification data indicates that there is a corresponding upgrade file, return an allowed upgrade response data and an incremental file index data; otherwise, return a canceled upgrade response data.

[0031] In an embodiment, the OTA upgrade operation comprises:

[0032] using the incremental file index data to download target firmware module data from the cloud, which contains program module identification;

[0033] According to the program module identifier, a specified program module to be upgraded in the UAV controller is located, and a firmware rewrite is performed on a storage partition corresponding to the specified program module.

[0034] In an embodiment, the method further comprises:

[0035] After performing the OTA upgrade operation, a verification flight test is performed at a preset safe height, and test flow data is collected;

[0036] A flow failure number is determined based on the test flow data.

[0037] When the flow failure number is greater than a preset second number, a version rollback operation is triggered to restore the original firmware.

[0038] In addition, to achieve the above-mentioned purpose, the present application also provides an OTA upgrade system based on UAV fault detection, which comprises a memory, a processor, and a UAV fault detection and OTA linkage upgrade program stored on the memory and executable on the processor, wherein the UAV fault detection and OTA linkage upgrade program is configured to implement the steps of the OTA upgrade method based on UAV fault detection.

[0039] The OTA upgrade method and system based on UAV fault detection provided by the present application can obtain real-time UAV operation data, generate preliminary fault data when the motor current flow failure number in the UAV operation data is greater than a preset first number in a single flight process, collect real-time auxiliary sensor data of the UAV through an on-board auxiliary sensor, obtain flight scene data of the UAV through an on-board GPS module, and combine the auxiliary sensor data, the flight scene data, and the preliminary fault data for fusion analysis to generate fault root cause probability data, so as to accurately identify software compatibility faults through multi-dimensional data fusion analysis, avoid false triggering of the upgrade operation, and improve the operation reliability of the UAV. BRIEF DESCRIPTION OF DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced here. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0042] Figure 1 The flowchart provided by an embodiment of the OTA upgrade method based on UAV fault detection of the present application is shown.

[0043] Figure 2 For this application Figure 1 A detailed flowchart of step S200;

[0044] Figure 3 This is a flowchart illustrating another embodiment of the OTA upgrade method based on UAV fault detection provided in this application;

[0045] Figure 4 This is a schematic diagram of a structural embodiment of the OTA upgrade system based on UAV fault detection provided in this application.

[0046] Explanation of icon numbers:

[0047] 10. Memory; 20. Processor.

[0048] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0049] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of this application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0050] It should be understood that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0051] In existing technologies, UAV control systems commonly employ threshold-based fault detection mechanisms, such as triggering firmware upgrade commands when the number of detected overcurrent faults exceeds a preset threshold. However, this method has significant drawbacks: it lacks a single fault diagnosis dimension and cannot distinguish between software compatibility faults and anomalies caused by hardware or environmental factors. For example, a momentary overcurrent caused by motor overheating might be incorrectly attributed to a software defect, thus triggering an unnecessary upgrade process. This not only wastes network and computing resources but may also affect operational efficiency due to task interruptions during the upgrade process.

[0052] To solve the above problems, a multi-dimensional data fusion analysis mechanism needs to be established to accurately identify the root cause of the fault, and the upgrade timing is dynamically scheduled in combination with the flight task state. By introducing auxiliary sensor data and flight scene data, hardware abnormalities and software compatibility problems can be effectively distinguished. Further, in combination with real-time control mode and throttle signal data, a task urgency identifier is generated, which can intelligently select the upgrade execution timing according to the work scene, avoiding interruption of critical tasks.

[0053] Based on this, the embodiment of the application provides an OTA upgrade method based on unmanned aerial vehicle fault detection, referring to Figure 1 , the OTA upgrade method based on unmanned aerial vehicle fault detection comprises steps S100-S500, wherein:

[0054] Step S100, real-time acquisition of unmanned aerial vehicle operation data, when the motor current overcurrent times in the unmanned aerial vehicle operation data in a single flight process is greater than a preset first number, generating preliminary fault data;

[0055] Step S200, real-time acquisition of auxiliary sensor data of the unmanned aerial vehicle by the on-board auxiliary sensor, acquisition of flight scene data of the unmanned aerial vehicle by the on-board GPS module, and fusion analysis of the auxiliary sensor data, the flight scene data and the preliminary fault data to generate fault root cause probability data; when the software compatibility problem probability in the fault root cause probability data exceeds a preset probability, generating a potential fault upgrade instruction;

[0056] Step S300, when the potential fault upgrade instruction or the user triggered active upgrade instruction is received, the fault root cause probability data and the unmanned aerial vehicle model data are sent to the cloud server for upgrade judgment processing;

[0057] Step S400, when the allow upgrade response data returned by the cloud server is received, an allow upgrade instruction is generated, and real-time control mode data and throttle signal data are acquired based on the allow upgrade instruction to generate task urgency identifier data;

[0058] Step S500, when the task urgency identifier data indicates that the unmanned aerial vehicle is in the take-off idle speed mode, the OTA upgrade operation is controlled to be executed after the unmanned aerial vehicle returns and lands; when the unmanned aerial vehicle is in the work mode, the OTA upgrade operation is delayed to the end of the work task.

[0059] In this embodiment, real-time acquisition of unmanned aerial vehicle operation data can be through continuous collection of current and speed parameters by on-board sensors. When the number of overcurrent events in a single flight exceeds a preset value, for example, 5 times, preliminary fault data generation is triggered. Auxiliary sensor data can include physical parameters collected by temperature sensors and accelerometers, and flight scene data can include altitude and geographic coordinate information. Fusion analysis can be performed by a weighted algorithm to associate temperature fluctuations, voltage abnormalities, and flight scene characteristics. For example, when a high-temperature environment is detected and the positioning data matches a preset conflict scenario, the probability of software compatibility problems is increased. Task urgency identification data can be based on throttle signals and control mode to determine the operation state. For example, when the throttle signal exceeds a preset threshold and is in automatic route mode, it is determined as a high-priority operation task.

[0060] In this embodiment, the motor current parameter is continuously monitored during flight. When the cumulative number of overcurrent events in a single flight cycle exceeds a preset threshold, preliminary fault judgment is triggered. Then, the motor temperature change detected by the temperature sensor, the voltage fluctuation data, and the flight height and geographic coordinate information obtained by the GPS are integrated to exclude environmental interference factors through multi-dimensional data analysis. For example, when the motor temperature is continuously higher than the safety threshold and is in a high-altitude area, the system will reduce the probability weight of software compatibility problems. After confirming that the software problem probability exceeds the preset threshold, the system decides on the upgrade opportunity according to the real-time flight state: if the unmanned aerial vehicle is in a hovering standby state, it immediately executes a return upgrade; if it is in a surveying, spraying, or other operation task, it delays the upgrade until the task is completed.

[0061] Compared with the prior art, the existing scheme only relies on a single overcurrent number to trigger an upgrade, while the present scheme achieves accurate identification of fault sources by fusing temperature, voltage, and scene data, effectively distinguishing between hardware abnormalities and software defects. Moreover, the prior art lacks task state awareness, while the present scheme introduces control mode and throttle signal analysis mechanisms, enabling dynamic scheduling of upgrade processes based on the urgency of the operation, thereby avoiding interruptions during critical tasks. Thus, the present application can reduce false upgrade operations caused by environmental factors and reduce network resource consumption. The task state awareness mechanism ensures operation continuity and avoids unnecessary interruptions. Multi-dimensional data analysis improves fault diagnosis accuracy, ensuring that upgrade operations are targeted at real software compatibility problems.

[0062] In a feasible implementation, after receiving the potential fault upgrade instruction or the user-triggered active upgrade instruction, sending the fault root cause probability data and the unmanned aerial vehicle model data to the cloud server for upgrade decision processing, the method further includes: when receiving the cloud returned cancel upgrade response data, generating a cancel upgrade instruction, and generating fault warning data based on the cancel upgrade instruction, uploading the fault warning data to the cloud server to perform a fault warning operation.

[0063] In this embodiment, the cancel upgrade instruction refers to the control signal sent by the cloud server to the UAV to terminate the upgrade process when it is determined that the upgrade is not needed at present. Specifically, it can be implemented by generating a response data packet containing a status code through a preset communication protocol, which is used to prevent triggering an invalid upgrade process when there is no available upgrade file or the scene does not match. The fault warning data refers to the combined information containing the fault root probability data and the UAV operating state. Specifically, it can be implemented by encapsulating temperature, voltage, flight scene parameters and fault mode similarity data in a structured data format, which is used to provide multi-dimensional basis for subsequent manual troubleshooting or automatic diagnosis. The fault warning operation refers to the process of classifying and storing the received warning data and marking the priority. Specifically, it can be implemented by associating the data to the historical fault records of the corresponding UAV model through a database interface, which is used to realize the cross-device synchronous warning and trend analysis of potential risks.

[0064] In this embodiment, when the cloud server determines that there is no matching incremental upgrade file according to the fault root probability data and the UAV model data, it returns a cancel upgrade response data to the UAV. After receiving the data, the UAV immediately terminates the upgrade process and generates a cancel upgrade instruction, which triggers the local warning module to integrate the current fault data to form fault warning data containing timestamp, environmental parameters and hardware state. After the data is uploaded to the cloud through an encrypted channel, it is automatically classified into a preset warning queue, and the cloud server starts the notification mechanism according to the queue priority, such as sending alarm information to the ground control station or updating the device health status list.

[0065] In this embodiment, the present scheme combines the cancel upgrade response with the warning data upload, so that the faults caused by software mismatch can be continuously tracked, and data support is provided for subsequent upgrade file development, avoiding the limitation of single device determination result affecting the overall diagnosis efficiency. The problem of interruption of fault handling process when there is no available upgrade file in the cloud is solved. By automatically generating and uploading warning data, the integrity of fault information can be maintained after canceling the upgrade, reducing the network resource consumption caused by repeated triggering of upgrade request, and providing data basis for subsequent batch fault analysis and targeted firmware development.

[0066] In a feasible implementation, before the step of sending the fault root probability data and the UAV model data to the cloud server for upgrade determination processing upon receiving the potential fault upgrade instruction or the user-triggered proactive upgrade instruction, the method further includes: extracting motor current time series data and control signal time series data from the UAV operating data; inputting the motor current time series data and the control signal time series data into a pre-trained prediction model to output fault occurrence probability data; and generating the potential fault upgrade instruction when the fault occurrence probability data exceeds a warning threshold.

[0067] In this embodiment, motor current time series data refers to the motor operating current change information recorded in time sequence, which can be achieved by periodic sampling and storage as time series data by a current sensor, and is used to reflect the dynamic change of motor load state. The control signal time series data refers to the time sequence record of the speed control instruction signal sent by the unmanned aerial vehicle controller to the motor, which can be extracted and stored by the internal bus data capture module of the controller, and is used to analyze the correlation between the control logic and the motor response. The prediction model refers to a multi-dimensional time series analysis algorithm trained by historical fault data, which can be constructed by using a long short-term memory neural network, and the potential fault can be predicted by identifying the abnormal correlation mode of the current signal and the control signal. The fault probability data refers to the software compatibility fault possibility quantitative value output by the model, which can be mapped to the 0-1 interval probability value by normalization processing, and is used to quantitatively evaluate the urgency of the upgrade requirement. The warning threshold refers to the critical probability value for triggering the upgrade instruction, which can be dynamically set according to different model reliability indicators, for example, when set to 0.75, it means that when the model determines that the fault probability exceeds 75%, the upgrade process is started.

[0068] In this embodiment, the current sensor records the motor operating current at a fixed sampling frequency to form time series data, and the internal bus recording module of the controller synchronously captures the PWM speed control instruction signal. After the two kinds of time series data are divided by a sliding window, they are input into a pre-trained neural network model. The model analyzes the time sequence correlation characteristics of current fluctuations and control instructions, identifies abnormal response patterns, and generates a firmware upgrade instruction when the probability value output by the model exceeds a preset threshold. For example, in the case of continuous delayed response of the motor to the instruction without reaching the overcurrent frequency threshold, the model can identify the abnormal control logic in advance and trigger a preventive upgrade.

[0069] In this embodiment, the present scheme analyzes the dynamic relationship between the control signal and the execution response through time series data modeling, which can identify potential software defects in advance, reduce the false upgrade rate caused by single threshold determination, and achieve accurate differentiation of fault types through multi-dimensional time series feature analysis. In the case of intermittent abnormality of the motor without reaching the overcurrent threshold, software compatibility problems can be found in advance to avoid fault accumulation and flight accidents. At the same time, the warning mechanism based on probability determination can flexibly schedule upgrade resources according to the risk level, reducing the impact of unnecessary firmware transmission on work tasks.

[0070] In a possible implementation, with reference to Figure 2 , step S200 includes steps S210-S240, wherein:

[0071] In step S210, the preliminary fault data is matched with the historical fault database to generate fault pattern similarity data.

[0072] In step S220, temperature data and voltage data in the auxiliary sensor data are extracted, and when the temperature data continuously exceeds a preset temperature threshold and the voltage data fluctuates beyond a preset fluctuation threshold, hardware anomaly identification data is generated.

[0073] In step S230, height data and positioning data in the flight scene data are extracted, and when it is determined based on the height data and the positioning data that the current flight scene conforms to a preset software conflict scene, scene compatibility identification data is generated.

[0074] In step S240, the fault mode similarity data, the hardware anomaly identification data, and the scene compatibility identification data are used to generate the fault root cause probability data through a preset rule.

[0075] In this embodiment, the fault mode similarity data refers to a quantitative index generated by comparing the matching degree of the current fault characteristics and historical fault cases, which can be specifically realized by using a pattern recognition algorithm to calculate the similarity of current waveform characteristics, and is used to identify repetitive fault modes. The hardware anomaly identification data refers to a judgment result generated by monitoring the abnormality of temperature and voltage, which can be specifically realized by using temperature sensor continuous sampling data and voltage fluctuation standard deviation calculation, and is used to distinguish between hardware aging or environmental factors causing anomalies. The scene compatibility identification data refers to a marker generated by analyzing flight height and positioning information to determine whether it is in a specific software conflict scene, which can be specifically realized by using geofencing technology and height threshold comparison algorithm, and is used to identify the adaptation problem of software version and specific scene. The preset rule generates the fault root cause probability data, which refers to a decision mechanism that fuses multiple dimensions of judgment results by weighting, which can be specifically realized by using a multi-factor linear regression model or a decision tree algorithm, and is used to comprehensively evaluate the possibility weight of different fault root causes.

[0076] In this embodiment, after detecting the preliminary fault, the system first performs pattern matching on the current motor current anomaly characteristics and the typical fault waveforms in the historical fault database, and calculates the fault mode similarity value. At the same time, the temperature sensor continuously collects temperature data in the motor compartment, and if the temperature value exceeds the safety threshold for a plurality of consecutive sampling periods and the battery voltage fluctuation amplitude exceeds the normal range, a hardware anomaly identification is triggered. The flight scene analysis module determines whether the current is in a geographical area or height layer prone to software version conflict based on the GPS positioning coordinates and flight height data, combined with a preset software conflict area database. Finally, the system inputs the three types of judgment results into a preset fusion analysis model, for example, different weight coefficients are used to assign probabilities to software compatibility problems, hardware faults, environmental interference, and other factors, to generate probability distribution data containing the possibility of each fault root cause.

[0077] In this embodiment, the scheme introduces historical failure mode matching, multi-sensor data cross-validation, and intelligent flight scene analysis to build a multi-dimensional fault diagnosis system. For example, in the case of motor overheating leading to instantaneous overcurrent, hardware anomaly identification and scene compatibility identification can effectively rule out the possibility of software problems, avoiding false triggering of the upgrade process. In this way, the application can accurately distinguish between software compatibility faults and abnormal phenomena caused by hardware or environmental factors, reducing the misjudgment rate. When the flight scene meets the software conflict characteristics and the hardware state is normal, the software problem is prioritized and the upgrade is triggered; when temperature anomalies or voltage fluctuations are detected, the possibility of software failure is ruled out to avoid invalid upgrade operations. This multi-dimensional joint diagnosis mechanism effectively solves the problem of false triggering of upgrade instructions caused by traditional single threshold determination, improving fault diagnosis accuracy and system reliability.

[0078] In a feasible implementation, the method further comprises: acquiring real-time weather severity data through a meteorological interface; and dynamically adjusting the preset probability of the software compatibility problem probability according to the weather severity data.

[0079] In this embodiment, the meteorological interface refers to a communication module for receiving external meteorological data, which can specifically establish a connection with a meteorological service platform using a wireless transmission protocol to obtain real-time wind speed, rainfall intensity, or lightning warning information. Weather severity data refers to a quantitative indicator reflecting the degree of influence of current meteorological conditions on the operation of the unmanned aerial vehicle, which can be generated by a preset weather level division rule, for example, mapping wind speed intervals into different severity levels. Dynamically adjusting the preset probability of the software compatibility problem probability refers to automatically correcting the threshold for triggering the upgrade according to changes in meteorological conditions, which can be achieved by establishing a mapping table of weather severity and probability correction coefficients, for example, reducing the preset probability in thunderstorm weather to avoid misjudgment.

[0080] In this embodiment, after detecting that the number of motor current overflows exceeds a preset threshold, the system acquires real-time weather data of the current flight area through the meteorological interface, such as wind speed reaching 8 levels or there being a thunderstorm warning. According to a preset weather level division rule, the current wind speed is mapped to a high severity level to generate corresponding weather severity data. Then, based on this data, a preset correction coefficient table is called to reduce the preset probability of the software compatibility problem probability from the default value of 60% to 40%. As a result, the system only triggers the upgrade instruction when the fault root probability exceeds the adjusted threshold, avoiding the false attribution of instantaneous overflows caused by bad weather to software problems.

[0081] Compared with the prior art, the existing scheme only relies on a fixed threshold to judge software compatibility problems, without considering the influence of meteorological conditions on the running state of hardware. For example, strong wind may cause motor load mutation to trigger overcurrent, but the traditional method still regards it as a software defect. The present scheme can effectively distinguish between environmental factors and software failures by introducing weather severity data to dynamically correct the probability threshold, thereby reducing the false upgrade operation caused by meteorological interference. In this way, the present application solves the problem of false judgment of software compatibility caused by environmental factors in the prior art, and reduces the triggering frequency of invalid upgrade operation. For example, in strong wind weather, the system reduces the preset probability to avoid attributing motor overcurrent to software defects, thereby reducing the firmware upgrade process caused by false judgment, ensuring the continuity of work tasks and saving network resources.

[0082] In a feasible implementation, the step of generating task urgency identification data based on the allowed upgrade instruction, real-time control mode data and throttle signal data includes: when the real-time control mode data is an automatic route mode, the throttle signal data exceeds a preset work threshold and reaches a preset time length, generating a work mode identification as the task urgency identification data; when the control mode data is a manual mode and the throttle signal data is lower than a preset hovering threshold, generating an idle speed mode identification as the task urgency identification data.

[0083] In the embodiment, the real-time control mode data refers to the flight control mode in which the unmanned aerial vehicle is currently located, which can be obtained through the state register of the flight control system, for example, in the automatic route mode, the navigation module generates waypoint instructions, and in the manual mode, the input signal of the remote controller is received. The throttle signal data refers to a physical quantity representing the intensity of power output, which can be quantified by the duty cycle of pulse width modulation signal, and is used to judge whether the unmanned aerial vehicle is in a high-load work state. The task urgency identification data refers to a classification label of the execution priority of the current task, which can be realized by binary coding, and the work mode identification corresponds to a high-priority task, and the idle speed mode identification corresponds to an interruptible task.

[0084] In the embodiment, if the throttle signal continuously exceeds the work threshold in the automatic route mode, it indicates that the unmanned aerial vehicle is executing high-intensity work tasks such as mapping and spraying, and at this time, the work mode identification is generated to delay the upgrade operation. If the throttle signal is lower than the hovering threshold in the manual mode, it indicates that the unmanned aerial vehicle is in landing preparation or low-power hovering state, and at this time, the idle speed mode identification is generated to allow immediate execution of the upgrade. The preset time length parameter can be set according to the power characteristics of different models, for example, for agricultural unmanned aerial vehicles, it can be set to 30 seconds to ensure that transient throttle fluctuations are excluded.

[0085] Compared with the prior art, the conventional method only triggers upgrade according to the number of overflows, without considering the relevance of the flight control mode and the power output state. For example, in automatic route operation, even if there is a software compatibility problem, the prior art will still force the interruption of the spraying task to perform upgrade, while the present scheme can accurately identify the key operation stage and delay unnecessary upgrade operation by analyzing the combined state of the control mode and the throttle signal. In this way, the present application realizes the dynamic matching of the unmanned aerial vehicle task scene and the upgrade timing, avoiding interruption of the flight task in the key operation stage. Specifically, in the agricultural plant protection scene, when the unmanned aerial vehicle is in the automatic route spraying state and the throttle maintains high load output, the system will automatically postpone the firmware upgrade until the operation is completed, ensuring the continuity of pesticide spraying. In the power inspection scene, if the unmanned aerial vehicle is in the hovering and photographing state, the upgrade operation can be immediately performed to eliminate potential software faults.

[0086] In a feasible implementation, the upgrade determination process includes: the cloud server uses the fault root cause probability data to match the incremental upgrade file library to generate file existence identification data; when the file existence identification data indicates that there is a corresponding upgrade file, return the upgrade permission response data and the incremental file index data; otherwise, return the upgrade cancellation response data.

[0087] In this embodiment, the incremental upgrade file library refers to a database storing firmware incremental update files divided by version or module, which can be implemented by using a hash table structure combined with version number index for fast retrieval of upgrade files associated with fault root cause probability data. The file existence identification data refers to a logical variable used to mark whether there is a matching upgrade file in the incremental upgrade file library, which can be implemented by using a Boolean value or a status code to determine whether the upgrade condition is met.

[0088] In this embodiment, after receiving the fault root cause probability data, the cloud server first parses the fault type code and software version information in the data, and matches them with the upgrade file metadata in the incremental upgrade file library. For example, if the fault root cause probability data indicates that the probability of software compatibility problem exceeds the preset threshold, the corresponding patch file is searched in the incremental upgrade file library according to the unmanned aerial vehicle model and the current firmware version number. If there is an upgrade file that meets the version iteration requirement, the upgrade permission response data is generated with the download path index of the file; if no valid file is found, the upgrade cancellation instruction is returned to avoid invalid upgrade request.

[0089] In this embodiment, the scheme terminates the invalid process in advance through file existence identification data, reduces network resource consumption, can effectively filter the failure scenarios of no available upgrade file, triggers the upgrade process only when it is confirmed that the corresponding repair file exists in the cloud, thereby reducing the invalid data transmission amount and shortening the upgrade decision response time. For example, when the software compatibility failure is reported by the unmanned aerial vehicle, if it is detected in the cloud that there is no adaptive version in the incremental file library, the upgrade operation is directly cancelled, thereby avoiding occupying the communication bandwidth and controller processing resources.

[0090] In a feasible implementation, the OTA upgrade operation includes: downloading target firmware module data from the cloud using the incremental file index data, the data containing program module identification; locating a specified program module to be upgraded in the unmanned aerial vehicle controller according to the program module identification, and performing firmware rewriting on a storage partition corresponding to the specified program module.

[0091] In this embodiment, the incremental file index data refers to index information for identifying the incremental upgrade file in the cloud server, which can be implemented by using a hash value or a version number mapping table. Through the data, the upgrade file matched with the current failure root cause can be quickly located. The program module identification refers to coding information for distinguishing different functional modules in the unmanned aerial vehicle controller, which can be implemented by using a hexadecimal address segment or a module name string. Through the identification, the flight control module or the communication module to be upgraded can be accurately identified. The storage partition refers to an independent storage area divided in the unmanned aerial vehicle controller, which can be implemented by using a physical sector or a logical partition table of a flash memory chip. Through partition isolation, the normal operation of other modules in the upgrade process can be avoided.

[0092] In this embodiment, after the cloud server confirms that there is an available upgrade file, the unmanned aerial vehicle system initiates a download request according to the returned incremental file index data, and only obtains a target firmware module data packet related to the failure. The data packet embeds the program module identification, and the system locates the corresponding storage partition in the controller after analyzing the identification, for example, the sector address range corresponding to the flight attitude calculation module. In the upgrade process, only the target partition is executed for firmware rewriting operation, and the other partitions remain unchanged. For example, when it is detected that there is a software compatibility problem in the image transmission module, the system only downloads the incremental file of the module and updates the corresponding partition, without full-disk overwriting the controller firmware.

[0093] In this embodiment, the scheme realizes on-demand downloading through an incremental file index, effectively reduces data transmission volume and storage space occupation in combination with modular partition upgrade technology, and only performs local update on specific modules, thereby significantly reducing the risk of system service interruption. In this way, the application solves the problems of resource waste and poor system stability in the existing unmanned aerial vehicle OTA upgrade process. Through incremental downloading and modular upgrade, transmission and storage of irrelevant data are avoided, and the upgrade time window is shortened. The rewriting mechanism of partition isolation is adopted to ensure the continuity of the function of non-upgrade modules and prevent flight control interruption caused by firmware update. Precise upgrade at the module level also reduces the probability of version compatibility errors and improves the success rate of fault repair.

[0094] In a feasible implementation, with reference to Figure 3 , the method further includes steps S610-S630, in which:

[0095] Step S610, after performing the OTA upgrade operation, performing a verification flight test at a preset safe height and collecting test overcurrent data;

[0096] Step S620, determining the number of overcurrent faults based on the test overcurrent data;

[0097] Step S630, when the number of overcurrent faults is greater than a preset second number, triggering a version rollback operation to restore the original firmware.

[0098] In this embodiment, the verification flight test refers to a short flight test performed at a safe height range after upgrade, which can be implemented by setting the flight height to 5-10 meters above ground in hovering or low-speed cruising mode, and is used to verify the stability of the system after upgrade. The test overcurrent data refers to the motor current data collected in real time during the verification flight, which can be obtained by an onboard current sensor at a sampling frequency of 10-20 times per second, and is used to detect whether there is an abnormal current phenomenon after upgrade. The number of overcurrent faults refers to the cumulative number of times that the motor current exceeds the rated value during the verification flight, which can be implemented by comparing the real-time current value with a preset threshold, for example, when the current exceeds 120% of the rated value, it is counted as an overcurrent event. The version rollback operation refers to restoring the firmware in the unmanned aerial vehicle controller to the version before upgrade, which can be implemented by backup partition switching or firmware image overlay, for example, by using a dual-partition storage architecture to realize fast switching.

[0099] Specifically, after completing the OTA upgrade, the UAV is controlled to perform a verification flight task at a preset safe height. During the verification flight, the on-board sensor continuously collects motor current data, and records an overcurrent event when detecting that the current exceeds a preset threshold. If the number of overcurrent events accumulated during the verification period reaches a preset number (for example, 3), it is determined that the upgraded firmware has compatibility problems, and a rollback mechanism is automatically triggered. During the rollback process, the flight control system switches to the backup storage partition to load the original firmware, and sends a version abnormality alarm information to the cloud.

[0100] In some embodiments, the verification flight test can be set to include a combination of multiple typical flight actions, such as a combination test sequence including hovering, climbing, and turning, and the duration of each action can be set to 10-30 seconds. The determination of the overcurrent data can use a dynamic threshold adjustment mechanism, such as compensating and calculating the current threshold according to the ambient temperature. The version rollback operation can be combined with flight state detection functions, such as automatically executing after the UAV lands or forcibly restoring the next time it is powered on.

[0101] In this embodiment, the scheme can effectively identify system abnormalities caused by software upgrades by adding a verification flight test link, combining dynamic overcurrent detection and automatic rollback mechanism. Compared with the traditional method which only relies on pre-upgrade diagnosis, the scheme extends the fault detection range to the post-upgrade stage, forming a complete closed-loop verification system, solving the technical problem of system stability decline caused by software upgrade, and avoiding the impact of false upgrades on flight safety. Through the cooperative mechanism of verification flight and automatic rollback, the system reliability is ensured while reducing the risk of secondary faults caused by firmware defects, ensuring the continuity of UAV operation tasks.

[0102] In the embodiments of the present application, the OTA upgrade method based on UAV fault detection acquires real-time UAV operation data, generates preliminary fault data when the number of motor current overflows in the UAV operation data during a single flight is greater than a preset first number, acquires real-time auxiliary sensor data of the UAV through an on-board auxiliary sensor, acquires flight scene data of the UAV through an on-board GPS module, and combines the auxiliary sensor data, the flight scene data, and the preliminary fault data for fusion analysis to generate fault root cause probability data, accurately identifies software compatibility faults through multi-dimensional data fusion analysis, avoids false triggering of upgrade operations, and improves UAV operation reliability.

[0103] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the OTA upgrade method based on UAV fault detection of the present application. Further simple transformations based on this technical concept are within the scope of protection of the present application.

[0104] The present application also provides an OTA upgrade system based on UAV fault detection, which is described with reference toFigure 4 The UAV fault detection based OTA upgrading system comprises a memory 10, a processor 20, and a UAV fault detection and OTA linkage upgrading program stored on the memory 10 and capable of running on the processor 20, wherein the UAV fault detection and OTA linkage upgrading program is configured to implement the steps of the UAV fault detection based OTA upgrading method.

[0105] The UAV fault detection based OTA upgrading system provided by the present application adopts the UAV fault detection based OTA upgrading method in the above embodiments, and can avoid false triggering of the upgrading operation and improve the reliability of UAV operation. Compared with the prior art, the UAV fault detection based OTA upgrading system provided by the present application has the same beneficial effects as the UAV fault detection based OTA upgrading method provided by the above embodiments, and other technical features in the UAV fault detection based OTA upgrading system are the same as the features disclosed in the above method embodiments, which will not be described here.

[0106] It should be understood that various parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0107] The above is merely specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any equivalent structural transformation made by using the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A method for OTA upgrade based on UAV fault detection, characterized in that, The method comprises: real-time acquisition of unmanned aerial vehicle operation data, generation of preliminary fault data when the number of motor current overflows in the unmanned aerial vehicle operation data during a single flight is greater than a preset first number; real-time acquisition of auxiliary sensor data of the unmanned aerial vehicle through an on-board auxiliary sensor, acquisition of flight scene data of the unmanned aerial vehicle through an on-board GPS module, and fusion analysis of the auxiliary sensor data, the flight scene data and the preliminary fault data to generate fault root cause probability data; generation of a potential fault escalation instruction when the probability of software compatibility problems in the fault root cause probability data exceeds a preset probability; sending the fault root cause probability data and unmanned aerial vehicle model data to a cloud server for escalation judgment processing when the potential fault escalation instruction or a user-triggered active escalation instruction is received; generating an escalation permission instruction when an escalation permission response data returned by the cloud server is received, and generating task urgency identification data based on real-time control mode data and throttle signal data obtained based on the escalation permission instruction; controlling the unmanned aerial vehicle to perform an OTA upgrade operation after returning to the ground when the task urgency identification data indicates that the unmanned aerial vehicle is in a take-off idle speed mode, and postponing the OTA upgrade operation until the end of a work task when the unmanned aerial vehicle is in a work mode. 2.The method for OTA upgrade based on UAV fault detection according to claim 1, wherein, After the step of sending the fault root cause probability data and unmanned aerial vehicle model data to a cloud server for escalation judgment processing when the potential fault escalation instruction or a user-triggered active escalation instruction is received, the method further comprises: generating a cancellation escalation instruction when a cancellation escalation response data returned by the cloud server is received, and generating fault warning data based on the cancellation escalation instruction, and uploading the fault warning data to the cloud server for fault warning operation. 3.The method for OTA upgrade based on UAV fault detection according to claim 1, wherein, Before the step of sending the fault root cause probability data and unmanned aerial vehicle model data to a cloud server for escalation judgment processing when the potential fault escalation instruction or a user-triggered active escalation instruction is received, the method further comprises: extracting motor current time series data and control signal time series data from the unmanned aerial vehicle operation data; inputting the motor current time series data and the control signal time series data into a pre-trained prediction model to output fault occurrence probability data; generating the potential fault escalation instruction when the fault occurrence probability data exceeds a warning threshold. 4.The method for OTA upgrade based on UAV fault detection according to claim 1, wherein, The step of fusion analysis of the auxiliary sensor data, the flight scene data and the preliminary fault data to generate fault root cause probability data comprises: matching the preliminary fault data with a historical fault database to generate fault mode similarity data; extracting temperature data and voltage data from the auxiliary sensor data, and generating hardware abnormality identification data when the temperature data continuously exceeds a preset temperature threshold and the voltage data fluctuates beyond a preset fluctuation threshold; extracting height data and positioning data from the flight scene data, and generating scene compatibility identification data when it is determined based on the height data and the positioning data that the current flight scene conforms to a preset software conflict scene; The fault mode similarity data, the hardware exception identification data and the scene compatibility identification data are generated into the fault root probability data through a preset rule. 5.The method for OTA upgrade based on UAV fault detection according to claim 1, wherein, The method further comprises: obtaining real-time weather severity data through a weather interface; adjusting the preset probability of the software compatibility problem probability according to the weather severity data. 6.The method for OTA upgrade based on UAV fault detection according to claim 1, wherein, The step of generating task urgency identification data based on the allowed upgrade instruction includes: generating a work mode identification as the task urgency identification data when the real-time control mode data is an automatic route mode, the throttle signal data exceeds a preset work threshold and reaches a preset time length; generating an idle speed mode identification as the task urgency identification data when the control mode data is a manual mode and the throttle signal data is lower than a preset hovering threshold. 7.The method for OTA upgrade based on UAV fault detection according to claim 1, wherein, The upgrade determination process includes: The cloud server matches the fault root probability data with an incremental upgrade file library to generate file existence identification data; when the file existence identification data indicates that there is a corresponding upgrade file, return an allowed upgrade response data and incremental file index data; otherwise, return a cancel upgrade response data. 8.The method for OTA upgrade based on UAV fault detection according to claim 7, wherein, The OTA upgrade operation includes: downloading target firmware module data from the cloud using the incremental file index data, which contains program module identification; locating a specified program module that needs to be upgraded in the unmanned aerial vehicle controller according to the program module identification, and performing firmware rewriting on the storage partition corresponding to the specified program module. 9.The method for OTA upgrade based on UAV fault detection of claim 1, wherein, The method further comprises: after performing the OTA upgrade operation, performing a verification flight test at a preset safe height and collecting test overcurrent data; determining the number of overcurrent faults based on the test overcurrent data; when the number of overcurrent faults is greater than a preset second number, triggering a version rollback operation to restore the original firmware.

10. An OTA upgrade system based on unmanned aerial vehicle fault detection, characterized in that, The OTA upgrade system based on unmanned aerial vehicle fault detection includes a memory, a processor and an unmanned aerial vehicle fault detection and OTA linkage upgrade program stored on the memory and executable on the processor, and the unmanned aerial vehicle fault detection and OTA linkage upgrade program is configured to implement the steps of the OTA upgrade method based on unmanned aerial vehicle fault detection as claimed in any one of claims 1 to 9.

Images