A method, system, device and medium for verifying the reliability of an unmanned aerial vehicle

Abstract

The present invention provides a method, system, device and medium for verifying the reliability of an unmanned aerial vehicle (UAV), including: obtaining environmental parameter data of the UAV to be verified under non-standard working conditions; calculating the battery efficiency of the UAV according to the environmental parameter data; using the Kalman filtering algorithm to estimate the flight state of the UAV based on the environmental parameter data and the battery efficiency, so as to obtain predicted flight state data; using a deep reinforcement learning algorithm to detect faults of the UAV according to the predicted flight state data, obtaining a fault risk index of the UAV, and verifying the reliability of the UAV according to the fault risk index; by introducing battery efficiency compensation and air density correction into the Kalman filtering algorithm in this application, the flight state of the UAV under non-standard working conditions can be predicted more accurately; by fusing state features, environmental parameters and battery efficiency through the deep reinforcement learning algorithm, it is beneficial to accurately calculate the fault risk index and improve the accuracy of UAV reliability verification.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV reliability verification, and particularly to a method, system, device and medium for verifying the reliability of UAVs. Background Art

[0002] At present, with the rapid development of Unmanned Aerial Vehicle (UAV) technology, UAVs are widely used in multiple fields such as agriculture, logistics, surveying and mapping, and military. The reliability verification of UAVs is considered a key factor to ensure their safe and efficient operation. Reliability verification can not only improve the stability and safety of UAVs in performing complex tasks, reduce operation risks, but also enhance users' trust in UAV technology, thus promoting the further development of the UAV market.

[0003] However, the reliability verification of UAVs is mainly based on standard working conditions (usually referring to controllable and relatively stable environmental conditions in a laboratory, including parameters such as temperature, humidity and air pressure within a conventional range). There are also significant limitations in the reliability verification based on standard working conditions. First of all, standard working conditions cannot fully reproduce various extreme and variable environments that may be encountered in actual operations. For example, in high-altitude areas, low temperature, thin air and rapidly changing meteorological conditions will significantly affect the power output and flight stability of UAVs, resulting in performance degradation and potentially leading to safety hazards. Secondly, the tests under standard working conditions often do not consider the problem of battery efficiency decline in extreme environments, which may lead to an incorrect assessment of the battery endurance, and further affect the entire operation period of the UAV. Finally, the research results under standard working conditions may not be applicable to actual application scenarios, making the evaluation of UAV reliability inaccurate in specific environments, thus affecting the success rate of tasks and flight safety.

[0004] Therefore, the existing UAV reliability verification method based on standard working conditions cannot fully consider the influence of extreme conditions such as high-altitude low temperature and thin air on the performance and battery endurance of UAVs under non-standard working conditions, resulting in inaccurate evaluation results, which may lead to safety hazards and mission failures. Summary of the Invention

[0005] In order to solve the problem that the existing UAV reliability verification method based on standard working conditions cannot fully consider the influence of extreme conditions under non-standard working conditions on the performance and battery endurance of UAVs, resulting in inaccurate reliability verification results, the present invention proposes a method for verifying the reliability of UAVs, including:

[0006] Obtaining environmental parameter data of the UAV to be verified under non-standard working conditions;

[0007] Calculating the battery efficiency of the UAV according to the environmental parameter data;

[0008] Based on the environmental parameter data of the drone and the battery efficiency, use the Kalman filtering algorithm to estimate the flight state of the drone, and obtain the predicted flight state data of the drone;

[0009] Based on the predicted flight state data of the drone, use the deep reinforcement learning algorithm to detect faults of the drone, obtain the fault risk index of the drone, and verify the reliability of the drone according to the fault risk index;

[0010] Among them, battery efficiency compensation and air density correction are introduced in the Kalman filtering algorithm.

[0011] Optionally, the step of using the Kalman filtering algorithm to estimate the flight state of the drone based on the environmental parameter data of the drone and the battery efficiency, and obtaining the predicted flight state data of the drone includes:

[0012] Based on the environmental parameter data of the drone and the battery efficiency, initialize the flight state of the drone to obtain the initial flight state of the drone;

[0013] Based on the environmental parameter data of the drone, the battery efficiency, and the initial flight state, perform state prediction on the drone through the state prediction equation to obtain the predicted flight state data of the drone.

[0014] Optionally, the expression corresponding to the state prediction equation is as follows:

[0015] ;

[0016] In the formula,

[0017] ;

[0018] ;

[0019] Among them, represents the predicted flight state data of the drone at time ; represents the state transition matrix of the drone at time ; represents the predicted flight state data of the drone at time ; represents the battery efficiency compensation factor; represents the control input matrix of the drone at time ; represents the control vector of the drone at time ; represents the battery efficiency of the drone; represents the process noise of the UAV at a certain moment ; represents the air density correction factor represents the UAV wind speed influence coefficient matrix of the UAV at a certain moment represents the wind speed vector represents the air density represents the standard air density

[0020] Optionally, the fault detection of the UAV is performed by using a deep reinforcement learning algorithm based on the predicted flight state data of the UAV, and the fault risk index of the UAV is obtained, including:

[0021] Feature extraction is performed according to the predicted flight state data of the UAV to obtain the state features of the UAV

[0022] According to the state features, the corresponding feature function values are calculated

[0023] Risk value mapping is performed according to the feature function values to obtain the fault risk index of the UAV

[0024] Among them, the state features include one or more of the following: position deviation, speed fluctuation, and attitude angle deviation

[0025] Optionally, the expression corresponding to the deep reinforcement learning algorithm is as follows:

[0026] ;

[0027] Among them, represents the fault risk index of the UAV represents the risk value mapping function represents the feature weight of the ; represents the total number of state features represents the battery efficiency of the UAV represents the feature vector of the state feature function value of represents the temperature influence coefficient represents the standard temperature represents the difference between the current temperature and the standard temperature represents the air density influence coefficient represents the current air density represents the standard air density represents the battery efficiency influence coefficient

[0028] Optionally, the calculation formula for the battery efficiency of the drone is as follows:

[0029] ;

[0030] Wherein, represents the battery efficiency of the drone; represents the reference value of the drone battery efficiency under standard operating conditions; represents the exponential function; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the load factor; represents the current battery current data of the drone; represents the maximum allowable current of the battery; represents the humidity influence coefficient; represents the current humidity data of the drone; represents the maximum allowable humidity of the drone.

[0031] Optionally, the environmental parameter data includes: environmental data and drone status data;

[0032] The environmental data includes one or more of the following: environmental temperature, air density, humidity data, wind speed data, and position data;

[0033] The drone status data includes one or more of the following: battery voltage data, battery current data, and sensor data.

[0034] Based on the same inventive concept, the present invention also provides a verification system for the reliability of a drone, including:

[0035] A data acquisition module for acquiring environmental parameter data of the drone to be verified under non-standard operating conditions;

[0036] An efficiency calculation module for calculating the battery efficiency of the drone according to the environmental parameter data;

[0037] A state estimation module for estimating the flight state of the drone using the Kalman filtering algorithm based on the environmental parameter data and the battery efficiency of the drone to obtain the predicted flight state data of the drone;

[0038] A fault detection module, which is used to perform fault detection on the drone according to the predicted flight state data of the drone by using a deep reinforcement learning algorithm, obtain a fault risk index of the drone, and perform reliability verification on the drone according to the fault risk index;

[0039] Among them, battery efficiency compensation and air density correction are introduced into the Kalman filtering algorithm.

[0040] Optionally, the state estimation module includes:

[0041] A state initialization sub-module, which is used to initialize the flight state of the drone according to the environmental parameter data of the drone and the battery efficiency to obtain the initial flight state of the drone;

[0042] A state prediction sub-module, which is used to perform state prediction on the drone through a state prediction equation according to the environmental parameter data of the drone, the battery efficiency, and the initial flight state to obtain the predicted flight state data of the drone.

[0043] Optionally, the expression corresponding to the state prediction equation is as follows:

[0044] ;

[0045] In the formula,

[0046] ;

[0047] ;

[0048] Among them, represents the predicted flight state data of the drone at time; represents the state transition matrix of the drone at time; represents the predicted flight state data of the drone at time; represents the battery efficiency compensation factor; represents the control input matrix of the drone at time; represents the control vector of the drone at time; represents the battery efficiency of the drone; represents the drone at time process noise; represents the air density correction factor; represents the drone at time wind speed influence coefficient matrix; represents the wind speed vector; represents the air density; represents the standard air density.

[0049] Optionally, the fault detection module includes:

[0050] A feature extraction sub-module for extracting features based on the predicted flight state data of the UAV to obtain the state features of the UAV;

[0051] A function value calculation sub-module for calculating the corresponding feature function value according to the state features;

[0052] A risk mapping sub-module for mapping the risk value according to the feature function value to obtain the fault risk index of the UAV.

[0053] Among them, the state features include one or more of the following: position deviation, speed fluctuation, and attitude angle deviation.

[0054] Optionally, the expression corresponding to the deep reinforcement learning algorithm is as follows:

[0055] ;

[0056] Among them, represents the fault risk index of the UAV; represents the risk value mapping function; represents the th feature weight of the state feature; ; represents the total number of state features; represents the battery efficiency of the UAV; represents the th state feature function value of the feature vector ; represents the temperature influence coefficient; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the battery efficiency influence coefficient.

[0057] Optionally, the calculation formula corresponding to the battery efficiency of the UAV is as follows:

[0058] ;

[0059] Among them, represents the battery efficiency of the UAV; represents the battery efficiency reference value of the UAV under standard working conditions; represents an exponential function; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the load factor; represents the current battery current data of the UAV; represents the maximum allowable current of the battery; represents the humidity influence coefficient; represents the current humidity data of the UAV; represents the maximum allowable humidity of the UAV.

[0060] Optionally, the environmental parameter data includes: environmental data and UAV status data;

[0061] The environmental data includes one or more of the following: environmental temperature, air density, humidity data, wind speed data, and position data;

[0062] The UAV status data includes one or more of the following: battery voltage data, battery current data, and sensor data.

[0063] On the other hand, the present invention also provides an electronic device, including: at least one processor and a memory; the memory and the processor are connected by a bus;

[0064] The memory is used to store one or more programs;

[0065] When the one or more programs are executed by the at least one processor, a method for verifying the reliability of a UAV as described above is implemented.

[0066] On the other hand, the present invention also provides a computer-readable storage medium with an execution program stored thereon. When the execution program is executed, a method for verifying the reliability of a UAV as described above is implemented.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] The present invention provides a method, system, device, and medium for verifying the reliability of an unmanned aerial vehicle (UAV), including: obtaining environmental parameter data of the UAV to be verified under non-standard working conditions; calculating the battery efficiency of the UAV according to the environmental parameter data; estimating the flight state of the UAV using the Kalman filter algorithm based on the environmental parameter data and the battery efficiency of the UAV to obtain predicted flight state data of the UAV; performing fault detection on the UAV using a deep reinforcement learning algorithm according to the predicted flight state data of the UAV to obtain a fault risk index of the UAV, and verifying the reliability of the UAV according to the fault risk index; wherein, battery efficiency compensation and air density correction are introduced into the Kalman filter algorithm; by introducing battery efficiency compensation and air density correction into the Kalman filter algorithm in this application, the flight state of the UAV under non-standard working conditions can be predicted more accurately; through the deep reinforcement learning algorithm, state features, environmental parameters, and battery efficiency can be fused, which is beneficial to accurately calculating the fault risk index and improving the accuracy of UAV reliability verification, thereby improving the safety and reliability of the UAV when performing tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is a schematic flow chart of a method for verifying the reliability of an unmanned aerial vehicle provided by the present invention;

[0070] Figure 2 is a schematic framework flow chart of flight state estimation in a method for verifying the reliability of an unmanned aerial vehicle provided by the present invention;

[0071] Figure 3 is a schematic structural composition diagram of a system for verifying the reliability of an unmanned aerial vehicle provided by the present invention;

[0072] Figure 4 is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] The present invention proposes a method, system, device, and medium for verifying the reliability of an unmanned aerial vehicle. The following further describes the specific embodiments of the present invention in detail with reference to the accompanying drawings.

[0074] Example 1:

[0075] The present invention provides a method for verifying the reliability of an unmanned aerial vehicle. The schematic flow chart is as Figure 1 shown, including:

[0076] Step 1: Obtain environmental parameter data of the UAV to be verified under non-standard working conditions;

[0077] Step 2: Calculate the battery efficiency of the UAV according to the environmental parameter data;

[0078] Step 3: Based on the environmental parameter data and battery efficiency of the unmanned aerial vehicle (UAV), use the Kalman filtering algorithm to estimate the flight state of the UAV and obtain the predicted flight state data of the UAV;

[0079] Step 4: Based on the predicted flight state data of the UAV, use the deep reinforcement learning algorithm to detect faults in the UAV, obtain the fault risk index of the UAV, and verify the reliability of the UAV according to the fault risk index;

[0080] Among them, battery efficiency compensation and air density correction are introduced in the Kalman filtering algorithm.

[0081] Generally, the reliability verification of UAVs is mainly carried out based on standard working conditions (such as ideal environments of normal temperature, normal pressure, no wind, etc.). This verification method evaluates the performance indicators of UAVs (such as flight stability, battery endurance, etc.) through laboratory simulation or limited on-site tests. However, under non-standard working conditions such as high altitude and low temperature (usually referring to the situation where the altitude is greater than 3000 meters and the environmental temperature is lower than 0°C), thin air (usually referring to the situation where the air density is lower than the air density at standard sea level), and strong wind (usually referring to the situation where the wind speed is significantly higher than the wind speed threshold under normal flight conditions of the UAV), the performance and battery endurance of the UAV may decrease significantly, and the existing methods cannot fully consider the influence of these extreme conditions, resulting in a large deviation between the reliability evaluation results and the actual situation. Therefore, the present invention conducts research by considering the environmental parameter data under non-standard working conditions (which can be collected by electronic devices embedded in the UAV); for example, the environmental parameter data in the above step 1 may include: environmental data and UAV state data;

[0082] The above environmental data may include one or more of the following: environmental temperature, air density, humidity data, wind speed data, and position data;

[0083] The above UAV state data may include one or more of the following: battery voltage data, battery current data, and sensor data.

[0084] On the basis of obtaining the environmental parameter data and UAV state data, in order to accurately evaluate the change in battery performance of the UAV under non-standard working conditions, it is necessary to further analyze the influencing factors of battery efficiency. These environmental data and state data together constitute the comprehensive data basis for the operation of the UAV, making the calculation of battery efficiency an important link in the next step of verifying the reliability of the UAV. For example, the calculation formula corresponding to the battery efficiency of the UAV in the above step 2 may be as follows:

[0085] ;

[0086] Wherein, represents the battery efficiency of the UAV; Represents the benchmark value of the UAV battery efficiency under standard operating conditions; Represents the exponential function; Represents the standard temperature; Represents the difference between the current temperature and the standard temperature; Represents the air density influence coefficient; Represents the current air density; Represents the standard air density; Represents the load factor; Represents the current battery current data of the UAV; Represents the maximum allowable current of the battery; Represents the humidity influence coefficient; Represents the current humidity data of the UAV; Represents the maximum allowable humidity of the UAV; In this example, by introducing multi-modal data fusion and comprehensively considering the effects of multiple factors such as temperature, air density, battery load, and humidity on the UAV battery efficiency, the comprehensiveness and accuracy of battery efficiency calculation can be significantly improved. Traditional battery efficiency calculations usually only consider a single factor (such as temperature or load) and cannot fully reflect the changes in battery performance under complex operating conditions. However, in this example, by introducing an exponential function and a linear correction factor, the calculation results of battery efficiency can be dynamically adjusted to dynamically adapt to changes in different environmental conditions, making it closer to the actual operating conditions, especially in complex and extreme operating environments. For example, through the temperature change rate Correct the influence of low temperature on battery chemical activity, and through the air density change rate Correct the influence of thin air at high altitudes on battery heat dissipation, and through the load factor Correct the influence of high load on battery internal resistance, and through the humidity change rate Correct the influence of high humidity on battery performance. This method of multi-modal data fusion can not only improve the comprehensiveness of the model but also significantly enhance the accuracy of battery efficiency calculation, providing more reliable data support for the assessment of the UAV's endurance and the prediction of flight states, thereby enhancing the UAV's adaptability and mission execution ability under non-standard operating conditions.

[0087] The battery efficiency calculated by the above calculation formula can provide relatively accurate data support for the performance evaluation of the UAV under different environmental conditions. Using these refined and quantified parameters, it is possible to achieve a detailed prediction of the UAV's flight dynamics. Based on this, it is possible to choose to estimate the UAV's flight state through the Kalman filtering method to obtain more accurate predicted flight state data. Specifically:

[0088] In one implementation, such as Figure 2As shown, the process of estimating the flight state of the drone using the Kalman filtering algorithm based on the environmental parameter data and battery efficiency of the drone in step 3 above to obtain the predicted flight state data of the drone may include:

[0089] Initialize the flight state of the drone according to the environmental parameter data and battery efficiency of the drone to obtain the initial flight state of the drone;

[0090] Perform state prediction on the drone through the state prediction equation according to the environmental parameter data, battery efficiency and initial flight state of the drone to obtain the predicted flight state data of the drone;

[0091] In this implementation method, by combining the environmental parameter data of the drone (such as fixed-wing drones, multi-rotor drones, hybrid drones, etc.) with the battery efficiency for state initialization, it is beneficial to improve the ability to handle the uncertainty of the initial state, making the initial flight state more accurate; at this time, using the recursive characteristics of the Kalman filtering algorithm, by integrating dynamic environmental factors and battery efficiency into state initialization, the algorithm obtains a more accurate initial state at the beginning, and allows real-time adjustment of the flight path and energy consumption plan of the drone under continuously updated environmental parameter conditions, which is beneficial to enhancing the response ability to complex environments, thereby improving flight safety and efficiency, and also showing unique advantages in reducing energy consumption, extending the endurance time and improving the task completion rate. Especially, its recursive estimation characteristics enable the prediction model to respond to changing environmental conditions in real time during flight, thus showing stronger adaptability. Through this optimization method, not only can the accuracy of flight state prediction be directly improved, but also the overall operation efficiency of the drone can be indirectly enhanced, significantly extending the flight time and reducing the energy consumption.

[0092] Exemplarily, the expression corresponding to the above state prediction equation may be as follows:

[0093] ;

[0094] In the formula,

[0095] ;

[0096] ;

[0097] Among them, represents the predicted flight state data of the drone at time ; represents the state transition matrix of the drone at time ; represents the battery efficiency compensation factor; Denote the control input matrix of the UAV at time; Denote the control vector of the UAV at time; Denote the battery efficiency of the UAV; Denote the UAV at the process noise at time; Denote the air density correction factor; Denote the UAV at the wind speed influence coefficient matrix at time; Denote the wind speed vector; Denote the air density; Denote the standard air density; In this example, the battery efficiency of the UAV is taken as a key factor. By introducing the battery efficiency compensation factor and battery efficiency, the state prediction can be dynamically adjusted in combination with the battery performance. This method is in contrast to the traditional prediction model that ignores the influence of battery efficiency on the flight state, which is conducive to greatly improving the modeling accuracy of energy consumption and providing basic support for optimizing the endurance and energy management; Secondly, an air density correction factor is introduced into the model. Through the wind speed influence coefficient matrix C k and the wind speed vector incorporate the dynamic change analysis of environmental parameters. Compared with the simple assumption of a constant environment, this method comprehensively considers the complex influence of the changing environment on the flight state, especially in scenarios with large changes in wind speed and air density, and has significant advantages. By adjusting the ratio of air density to standard air density, the influence of air resistance can be effectively corrected, so as to more accurately predict the flight behavior of the UAV; In addition, in the consideration of process noise, noise interference can be effectively reduced, thus further ensuring the reliability and stability of flight prediction.

[0098] In the above steps, the flight state of the UAV is predicted by the Kalman filter algorithm, and relatively accurate predicted flight state data is obtained. However, relying solely on the prediction of the flight state is not sufficient to comprehensively evaluate the health status of the UAV. In order to further ensure the safe flight of the UAV in a complex environment, it is necessary to deeply analyze the predicted flight state data to identify potential fault risks. Specifically:

[0099] In one implementation, in step 4 above, according to the predicted flight state data of the UAV, the process of using the deep reinforcement learning algorithm to detect faults in the UAV and obtain the fault risk index of the UAV may include:

[0100] Extract features from the predicted flight state data of the UAV to obtain the state features of the UAV;

[0101] Calculate the corresponding feature function values according to the state features;

[0102] Perform risk value mapping based on the characteristic function value to obtain the fault risk index of the UAV;

[0103] Among them, the state characteristics include one or more of the following: position deviation, speed fluctuation, and attitude angle deviation;

[0104] Exemplarily, assume that the state characteristics of the UAV are , where, represents the position deviation; represents the speed fluctuation; represents the attitude angle deviation; Define a comprehensive feature vector :

[0105] ;

[0106] According to the feature vector , construct the characteristic function , and use the neural network model to calculate the characteristic function value:

[0107] ;

[0108] Among them, represents the neural network model; are the model parameters;

[0109] According to the characteristic function value perform risk value mapping, define a risk mapping function , and convert it into the fault risk index ;

[0110] In the risk mapping function, non-linear mapping can be selected to improve the sensitivity to changes in state characteristics. For example, the sigmoid function is used for normalization:

[0111] ;

[0112] Among them, is the slope of the mapping; is the bias term of the mapping, used to adjust the sensitivity and range of the risk index;

[0113] In this implementation method, through the accurate extraction of the UAV state characteristics, it can be ensured that the fault detection algorithm can be analyzed based on reliable data input. For example, in the example, the state characteristics can include the position deviation , the speed fluctuation , and the attitude angle deviation . These important state indicators are represented by the comprehensive feature vector Can be effectively integrated, which can greatly improve the accuracy of fault detection, and by using the neural network model to construct the feature function The process means that the model can automatically optimize and identify complex fault patterns by learning historical data. Compared with traditional methods, this implementation method can handle higher-dimensional and non-linear risks, making the determination of fault risks more accurate; further, the risk mapping function The adoption, especially the choice of non-linear mapping, can significantly improve the sensitivity to changes in state features, making the fault risk index Can more sensitively reflect environmental changes and state anomalies; therefore, this implementation method can improve the accuracy and sensitivity of fault detection through deep reinforcement learning, and by comprehensively considering state features, improve the overall adaptability of the UAV.

[0114] For example, the expression corresponding to the above deep reinforcement learning algorithm can be as follows:

[0115] ;

[0116] Among them, Represents the fault risk index of the UAV; Represents the risk value mapping function; Represents the Feature weight of the th state feature; ; Represents the total number of state features; Represents the battery efficiency of the UAV; Represents the Feature vector of the th state feature Value of the state feature function; Represents the temperature influence coefficient; Represents the standard temperature; Represents the difference between the current temperature and the standard temperature; Represents the air density influence coefficient; Represents the current air density; Represents the standard air density; Represents the battery efficiency influence coefficient; in this example, by introducing the method of weighted summation, the influences of multiple state features of the UAV (such as position deviation, speed fluctuation, and attitude angle deviation) are integrated, and the influence degree of different features is dynamically adjusted through the weight parameter This design enables the model to flexibly adjust the weights of features based on specific tasks or flight environments, significantly enhancing the adaptability and generalization ability of the model; in addition, multiple key environmental parameters (such as temperature difference , air density , standard density ), and the system performance parameters of the drone (such as battery efficiency ), through the temperature influence coefficient γ, the air density influence coefficient α, and the battery efficiency influence coefficient These variables are processed by hierarchical weighting. This design significantly improves the model's ability to handle complex environmental variables, and can dynamically capture the impacts of temperature changes, air density fluctuations, and battery performance degradation on potential drone failures. Especially in extreme environments or long-term flight scenarios, this method demonstrates stronger robustness and early warning capabilities. Finally, through the introduction of the risk value mapping function, the calculation results in the multivariate complex model can be mapped into an intuitive and quantifiable risk index R. This not only facilitates risk transmission and understanding in engineering practice, but also provides a clear guiding logic for the automatic decision-making and fault handling of drones.

[0117] In summary, the present invention aims at the problem that the existing drone reliability verification method based on standard working conditions cannot fully consider the impacts of extreme conditions such as high altitude, low temperature, and thin air in non-standard working conditions on the drone performance and battery endurance, resulting in inaccurate reliability verification results. A drone reliability verification method is proposed. By obtaining environmental parameter data under non-standard working conditions and calculating the battery efficiency of the drone, key environmental factors and the drone's own performance are introduced into the reliability analysis framework, breaking through the defect of being limited to the standard working condition assumption in traditional methods. The Kalman filter algorithm is used to estimate the flight state of the drone, and at the same time, combined with the battery efficiency compensation factor and air density correction, the predicted flight state data can more truly reflect the impact of non-standard working conditions on the drone. Further, through the deep reinforcement learning algorithm, the predicted flight state data is used for fault detection, dynamically capturing the risks that may occur to the drone in extreme environments, and quantitatively outputting the fault risk index. Finally, through the analysis and processing of the risk index, the comprehensive, accurate, and dynamic verification of the drone performance and reliability is realized. The method of the present invention effectively makes up for the deficiencies of the prior art in complex environments, significantly improves the reliability verification accuracy of drones under various non-standard working conditions, provides important technical support for the safe operation of drones in extreme mission scenarios, and also provides a scientific basis for the design optimization and operation management of drones.

[0118] Embodiment 2:

[0119] In practical applications, drones often need to perform tasks in complex and dynamic environments. Therefore, by combining fault detection with multi-aircraft collaborative tasks, it can be ensured that the entire task can still proceed smoothly when a single aircraft fails, thereby reducing the risk of task interruption. Specifically:

[0120] First, a shared state information platform can be established so that each drone can obtain the flight status and fault risk index of other drones. This information sharing can enhance the fault warning ability of the entire group and reduce the impact of single-point failures on the mission.

[0121] In the implementation of multi-drone collaborative missions, intelligent scheduling can be combined with the fault detection results of drones. For example, for drones with a high risk level, other drones with low risk can be mobilized to undertake their tasks to ensure the continuous execution of the overall mission. And based on real-time fault detection data, the mission plan can be dynamically adjusted. If a drone fails, it can be quickly identified and the mission can be automatically re-planned for other drones to minimize mission delays or risks.

[0122] Utilize the data collection in multi-drone collaboration to identify the fault modes jointly faced by multiple drones through a deep learning model, thereby optimizing the fault detection algorithm. At the same time, considering the structural and mission characteristics of different drones, the model has stronger adaptability.

[0123] And in multi-drone missions, a real-time feedback mechanism can also be considered to ensure that drones can identify and respond to faults with each other. For example, if a drone detects a fault, it can immediately transmit the information to other drones to initiate emergency measures.

[0124] Therefore, associating the implementation method of this application with multi-drone collaborative missions can not only improve the overall performance and safety of the drone system, but also effectively respond to the development direction of the drone industry.

[0125] Embodiment 3:

[0126] Based on the same inventive concept, the present invention also provides a verification system for the reliability of drones. The schematic structural composition is as Figure 3 shown, including:

[0127] A data acquisition module for acquiring environmental parameter data of the drone to be verified under non-standard working conditions.

[0128] An efficiency calculation module for calculating the battery efficiency of the drone according to the environmental parameter data.

[0129] A state estimation module for using the Kalman filter algorithm to estimate the flight state of the drone based on the environmental parameter data and battery efficiency of the drone, and obtaining the predicted flight state data of the drone.

[0130] A fault detection module for using the deep reinforcement learning algorithm to detect faults of the drone based on the predicted flight state data of the drone, obtaining the fault risk index of the drone, and verifying the reliability of the drone according to the fault risk index.

[0131] Among them, battery efficiency compensation and air density correction are introduced into the Kalman filter algorithm.

[0132] Exemplarily, the above environmental parameter data may include: environmental data and UAV state data;

[0133] The above environmental data may include one or more of the following: environmental temperature, air density, humidity data, wind speed data, and position data;

[0134] The above UAV state data may include one or more of the following: battery voltage data, battery current data, and sensor data.

[0135] Exemplarily, the calculation formula corresponding to the battery efficiency of the above UAV may be as follows:

[0136] ;

[0137] Wherein, represents the battery efficiency of the UAV; represents the battery efficiency reference value of the UAV under standard working conditions; represents the exponential function; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the load factor; represents the current battery current data of the UAV; represents the maximum allowable current of the battery; represents the humidity influence coefficient; represents the current humidity data of the UAV; represents the maximum allowable humidity of the UAV.

[0138] In one implementation manner, the above state estimation module may include:

[0139] A state initialization sub-module, configured to initialize the flight state of the UAV according to the environmental parameter data and battery efficiency of the UAV, and obtain the initial flight state of the UAV;

[0140] A state prediction sub-module, configured to perform state prediction on the UAV according to the environmental parameter data, battery efficiency, and initial flight state of the UAV, and obtain the predicted flight state data of the UAV.

[0141] Exemplarily, the expression corresponding to the above state prediction equation may be as follows:

[0142] ;

[0143] In the formula,

[0144] ;

[0145] ;

[0146] Among them, represents the predicted flight state data of the UAV at moment; represents the state transition matrix of the UAV at moment; represents the predicted flight state data of the UAV at moment; represents the battery efficiency compensation factor; represents the control input matrix of the UAV at moment; represents the control vector of the UAV at moment; represents the battery efficiency of the UAV; represents the process noise of the UAV at moment; represents the air density correction factor; represents the wind speed influence coefficient matrix of the UAV at moment; represents the wind speed vector; represents the air density; represents the standard air density.

[0147] In one implementation, the above-mentioned fault detection module may include:

[0148] A feature extraction sub-module, configured to extract features based on the predicted flight state data of the UAV to obtain the state features of the UAV;

[0149] A function value calculation sub-module, configured to calculate the corresponding feature function value according to the state features;

[0150] A risk mapping sub-module, configured to perform risk value mapping according to the feature function value to obtain the fault risk index of the UAV.

[0151] Among them, the state features include one or more of the following: position deviation, speed fluctuation, and attitude angle deviation.

[0152] Exemplarily, the expression corresponding to the above-mentioned deep reinforcement learning algorithm may be as follows:

[0153] ;

[0154] Among them, represents the fault risk index of the UAV; represents the risk value mapping function; represents the feature weight of the th state feature; represents the total number of state features; represents the battery efficiency of the drone; represents the th state feature function value of the state feature of the represents the temperature influence coefficient; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the battery efficiency influence coefficient.

[0155] Embodiment 4:

[0156] As Figure 4 shown, the present invention further provides an electronic device, which may be a computer device, a single-chip microcomputer device, a smart mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected through a bus; the memory can be used to store an execution program, and the exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory can also be used to store data, and the data can be called and / or modified when the instructions are executed.

[0157] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a method for verifying the reliability of a drone in the above embodiment.

[0158] Embodiment 5:

[0159] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device-readable storage medium (Memory). The electronic device-readable storage medium is a memory device in an electronic device, used to store programs and data. It can be understood that the storage medium here can include both the built-in storage medium in the electronic device and, of course, the extended storage medium supported by the electronic device. The storage medium provides a storage space, and this storage space stores the operating system of the terminal. And, in this storage space, there is also stored one or more instructions suitable for being loaded and executed by the processor. These instructions can be one or more executable programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. By the processor loading and executing one or more instructions stored in the storage medium, the steps of a method for verifying the reliability of a drone in the above embodiments can be implemented.

[0160] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0161] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0162] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and this instruction device realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for realizing the functions specified in one process or multiple processes and / or one block or multiple blocks in the flow Figure 1 one process or multiple processes and / or blocks Figure 1 or steps for realizing the functions specified in one block or multiple blocks.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the scope of its protection. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that after reading the present invention, various changes, modifications or equivalent replacements can still be made to the specific implementation manners of the application. However, these changes, modifications or equivalent replacements are all within the scope of the protection of the claims pending for the application.

Claims

1. A method for verifying the reliability of an unmanned aerial vehicle, characterized in that, including: Obtaining environmental parameter data of the UAV to be verified under non-standard working conditions; Calculating the battery efficiency of the UAV according to the environmental parameter data; Initializing the flight state of the UAV according to the environmental parameter data and the battery efficiency of the UAV to obtain the initial flight state of the UAV; Performing state prediction on the UAV through a state prediction equation according to the environmental parameter data, the battery efficiency and the initial flight state of the UAV to obtain the predicted flight state data of the UAV; Performing fault detection on the UAV by using a deep reinforcement learning algorithm according to the predicted flight state data of the UAV to obtain the fault risk index of the UAV, and performing reliability verification on the UAV according to the fault risk index; Wherein, battery efficiency compensation and air density correction are introduced into the state prediction equation; the environmental parameter data includes: environmental data and UAV state data; the environmental data includes: environmental temperature, air density, humidity data, wind speed data and position data; the UAV state data includes: battery voltage data, battery current data and sensor data; The calculation formula corresponding to the battery efficiency of the UAV is as follows: ； Among them, represents the battery efficiency of the drone; represents the reference value of the drone battery efficiency under standard working conditions; represents the exponential function; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the load factor; represents the current battery current data of the drone; represents the maximum allowable current of the battery; represents the humidity influence coefficient; represents the current humidity data of the drone; represents the maximum allowable humidity of the drone; The expression corresponding to the state prediction equation is as follows: ； In the formula, ； ； Among them, represents the predicted flight state data of the UAV at moment; represents the state transition matrix of the UAV at moment; represents the predicted flight state data of the UAV at moment; represents the battery efficiency compensation factor; represents the control input matrix of the UAV at moment; represents the control vector of the UAV at moment; represents the battery efficiency of the UAV; represents the process noise of the UAV at moment; represents the air density correction factor; represents the wind speed influence coefficient matrix of the UAV at moment; represents the wind speed vector; represents the air density; represents the standard air density.

2. The method according to claim 1, wherein The performing fault detection on the UAV by using a deep reinforcement learning algorithm according to the predicted flight state data of the UAV to obtain the fault risk index of the UAV includes: Performing feature extraction according to the predicted flight state data of the UAV to obtain the state features of the UAV; Calculating the corresponding feature function value according to the state features; Performing risk value mapping according to the feature function value to obtain the fault risk index of the UAV; Wherein, the state features include one or more of the following: position deviation, speed fluctuation and attitude angle deviation.

3. The method according to claim 1, characterized in that The expression corresponding to the deep reinforcement learning algorithm is as follows: ； Among them, represents the fault risk index of the UAV; represents the risk value mapping function; represents the characteristic weight of the th state characteristic; represents the total number of state characteristics; represents the battery efficiency of the UAV; represents the th state characteristic vector of the state characteristic function value; represents the temperature influence coefficient; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the battery efficiency influence coefficient.

4. A verification system for the reliability of an unmanned aerial vehicle, characterized in that, including: A data acquisition module, configured to obtain environmental parameter data of the UAV to be verified under non-standard working conditions; An efficiency calculation module, configured to calculate the battery efficiency of the UAV according to the environmental parameter data; A state estimation module, configured to perform flight state estimation on the UAV by using the Kalman filter algorithm according to the environmental parameter data and the battery efficiency of the UAV to obtain the predicted flight state data of the UAV; A fault detection module, configured to perform fault detection on the UAV by using a deep reinforcement learning algorithm according to the predicted flight state data of the UAV to obtain the fault risk index of the UAV, and perform reliability verification on the UAV according to the fault risk index; Wherein, battery efficiency compensation and air density correction are introduced into the Kalman filter algorithm; the environmental parameter data includes: environmental data and UAV state data; the environmental data includes: environmental temperature, air density, humidity data, wind speed data and position data; the UAV state data includes one or more of the following: battery voltage data, battery current data and sensor data; The calculation formula corresponding to the battery efficiency of the UAV is as follows: ； Among them, represents the battery efficiency of the drone; represents the reference value of the drone's battery efficiency under standard operating conditions; represents the exponential function; represents the standard temperature; represents the difference between the current temperature and the standard temperature; represents the air density influence coefficient; represents the current air density; represents the standard air density; represents the load factor; represents the current battery current data of the drone; represents the maximum allowable current of the battery; represents the humidity influence coefficient; represents the current humidity data of the drone; represents the maximum allowable humidity of the drone; The state estimation module includes: A status initialization sub-module, configured to initialize the flight status of the drone according to the environmental parameter data and the battery efficiency of the drone, so as to obtain the initial flight status of the drone; A status prediction sub-module, configured to perform status prediction on the drone through a status prediction equation according to the environmental parameter data, the battery efficiency and the initial flight status of the drone, so as to obtain the predicted flight status data of the drone; The expression corresponding to the status prediction equation is as follows: ； In the formula, ； ； Among them, represents the predicted flight state data of the UAV at moment; represents the state transition matrix of the UAV at moment; represents the predicted flight state data of the UAV at moment; represents the battery efficiency compensation factor; represents the control input matrix of the UAV at moment; represents the control vector of the UAV at moment; represents the battery efficiency of the UAV; represents the process noise of the UAV at moment; represents the air density correction factor; represents the wind speed influence coefficient matrix of the UAV at moment; represents the wind speed vector; represents the air density; represents the standard air density.

5. An electronic device, characterized in that, including: at least one processor and a memory; the memory and the processor are connected by a bus; the memory is configured to store one or more programs; when the one or more programs are executed by the at least one processor, the method for verifying the reliability of a drone as described in any one of claims 1 to 3 is implemented.

6. A computer-readable storage medium, characterized in that, There is an execution program stored thereon, and when the execution program is executed, the method for verifying the reliability of a drone as described in any one of claims 1 to 3 is implemented.

Images