A UAV control method and system

Abstract

The present invention relates to the field of unmanned aerial vehicle control technology, and in particular to a method and system for controlling an unmanned aerial vehicle. The method comprises the following steps: collecting flight environment data and flight status data in real time through sensors carried by the unmanned aerial vehicle. First, extracting wind condition and rainfall characteristic data in the flight environment; then, using the wind condition data to evaluate the structural fluid load of the unmanned aerial vehicle, and based on this, performing force imbalance structural analysis on the rainfall characteristic data; then, adjusting the attitude control of the unmanned aerial vehicle according to the analysis results, and performing logic learning of the control strategy based on the adjustment data; finally, designing the obtained unmanned aerial vehicle control logic strategy firmware and embedding it into the control center to realize the automatic execution of the flight control method. The present invention makes the unmanned aerial vehicle control technology more perfect by optimizing the unmanned aerial vehicle control technology.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle control, and in particular to a method and system for controlling an unmanned aerial vehicle. Background Art

[0002] With the rapid development of UAV technology, especially its wide application in agriculture, logistics, environmental monitoring, disaster relief and other fields, the stability and adaptability of UAVs in complex flight environments have received increasing attention. The flight stability of UAVs is not only affected by meteorological factors (such as wind speed, wind direction, rainfall, etc.), but is also closely related to its design structure and the real-time response capability of the flight control system. In practical applications, UAVs often need to cope with changeable and unpredictable flight environments, such as strong winds, sudden rainfall, airflow disturbances, etc., which have a negative impact on their flight attitude, heading control and mission execution. Therefore, how to dynamically adjust the control strategy during flight to ensure the flight stability and safety of UAVs under complex and severe weather conditions has become an important topic in UAV technology research. However, traditional UAV control methods have problems such as inaccurate analysis of wind conditions and rainfall, and large errors in UAV flight attitude control. Summary of the invention

[0003] Based on this, it is necessary to provide a drone control method and system to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for controlling a drone is provided, the method comprising the following steps:

[0005] Step S1: collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, and obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, and obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively;

[0006] Step S2: Perform structural fluid load evaluation on the UAV flight status data according to the real-time wind condition data of the flight environment to obtain structural fluid load data; perform force imbalance structural decomposition on the flight environment rainfall characteristic data based on the structural fluid load data to obtain rain impact force imbalance decomposition data;

[0007] Step S3: adjusting the attitude control of the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data; performing UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy;

[0008] Step S4: Automated execution firmware design is performed on the drone control logic strategy to obtain the drone control logic strategy firmware, and the drone control logic strategy firmware is embedded into the drone control center to execute the drone control method.

[0009] The present invention provides comprehensive basic information for subsequent analysis by collecting flight environment data and flight status data in real time through sensors carried by unmanned aerial vehicles. By extracting wind condition data and rainfall characteristic data, the dynamic changes of the flight environment can be fully understood, especially the real-time fluctuations of factors such as wind speed, wind direction and rainfall. The extraction of these data provides an accurate basis for the evaluation and adjustment of the subsequent flight status, ensuring that the environmental factors during the flight process can be fed back to the unmanned aerial vehicle system in a timely manner. According to the real-time wind condition data of the flight environment, the structural fluid load of the flight status data of the unmanned aerial vehicle is evaluated, and the specific impact of wind force and flight status on the structure of the unmanned aerial vehicle can be quantified, which helps to judge the potential flight risks. Further, these load data are used to analyze the force imbalance caused by rainfall, and the specific impact of rain on the flight stability of the unmanned aerial vehicle is identified through the force imbalance structural decomposition of the rainfall characteristic data, ensuring that the rain impact in flight can be timely identified and effectively dealt with. By combining the rain impact force imbalance decomposition data with the structural fluid load data, an accurate adjustment basis is provided for the attitude control system of the unmanned aerial vehicle. By dynamically adjusting the flight attitude, the negative impact of environmental changes on the stability of the unmanned aerial vehicle is effectively reduced. On this basis, the attitude control adjustment of the flight state is carried out, combined with the optimization of the flight strategy, to achieve intelligent adjustment and optimization during the flight process. The control strategy of the UAV is continuously improved through this process, which improves the safety and efficiency of the flight. The optimized UAV control logic strategy is converted into an automated execution firmware and embedded into the UAV control center to ensure that the control strategy can be executed in real time and automatically. The automated execution of the firmware design not only improves the autonomous flight capability of the UAV, but also reduces the dependence on human operation. By applying the optimized flight control logic firmware to the UAV control system, the UAV can adaptively adjust according to environmental changes and flight status, improve the reliability and stability of the flight, and ensure the smooth flight of the UAV in complex environments.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: collecting real-time data of the drone's flight environment and flight status through sensors carried by the drone, and obtaining real-time data of the drone's flight environment and flight status data;

[0012] Step S12: Fill missing values ​​in the real-time data of the UAV flight environment to obtain real-time filled data of the flight environment;

[0013] Step S13: extracting real-time wind condition data and rainfall characteristic data from the real-time filled data of the flight environment, and obtaining real-time wind condition data of the flight environment and rainfall characteristic data of the flight environment respectively.

[0014] The present invention collects real-time data of the flight environment and flight status through sensors carried by the drone, which is the basis of the whole process. The flight environment data acquired by the sensor in real time include factors such as wind speed, temperature, and air pressure, while the flight status data covers key information such as the attitude, speed, and altitude of the drone. These data provide comprehensive and accurate support for subsequent analysis and decision-making, ensuring that the environment and flight status during the flight process can be monitored and fed back in real time, and are the primary step in optimizing the flight control strategy. Because the drone encounters signal interference or sensor failure during flight, some environmental data is lost. This step restores the missing data through a suitable missing value filling method (such as interpolation, mean filling, etc.), thereby ensuring the integrity and accuracy of the data. This step is crucial for subsequent data analysis. The filled data can ensure the accuracy of subsequent calculations and decisions, and avoid flight risk assessment deviations caused by missing data. The flight environment data with missing value filling is further analyzed to extract wind condition and rainfall characteristic data. Wind condition data usually includes parameters such as wind speed and wind direction, which have a direct impact on flight stability; rainfall characteristic data reflects factors such as precipitation and precipitation intensity, which affect the flight control and structural bearing capacity of the drone. By extracting these key data, more precise parameters can be provided for real-time monitoring of the flight environment, which helps to determine the impact of environmental changes on flight stability and provide a basis for subsequent flight status assessment and attitude control.

[0015] Preferably, step S2 comprises the following steps:

[0016] Step S21: Acquire the morphological structure data of the drone;

[0017] Step S22: performing structural fluid load evaluation on the UAV flight state data and the UAV morphological structure data according to the real-time wind condition data of the flight environment to obtain structural fluid load data;

[0018] Step S23: Calculate the difference of local rain impact on the UAV based on the structural fluid load data and the rainfall characteristic data of the flight environment to obtain local rain impact difference data;

[0019] Step S24: performing force imbalance structural decomposition on the UAV flight state data and the UAV morphological structure data according to the local rain impact difference number to obtain rain impact force imbalance decomposition data.

[0020] The present invention first obtains the morphological structure data of the drone, including the geometric dimensions, mass distribution, material properties, etc. of each component of the drone. These data are the basis for evaluating the drone's ability to withstand various external loads in a complex flight environment. The morphological structure of the drone determines its flight performance and carrying capacity, so accurately obtaining these data is crucial for subsequent analysis. Only through accurate structural data can a reasonable flight state evaluation and control strategy design be carried out to ensure that the drone can safely cope with various external environmental changes during flight. According to the real-time wind condition data of the flight environment, the flight state data and morphological structure data of the drone are evaluated for structural fluid loads. This process calculates the structural loads under different flight states by simulating the interaction between wind force and the drone structure. Wind condition data such as wind speed and wind direction will directly affect the aerodynamic load of the drone, and flight state data such as speed and attitude will determine the magnitude and direction of the structural force. Through this evaluation, the aerodynamic load borne by the drone during flight can be quantified, providing a scientific basis for subsequent flight control and stability analysis, and ensuring that the drone can maintain flight stability under various wind conditions. Based on the calculated structural fluid load data, combined with the rainfall characteristic data in the flight environment, the local rain impact difference calculation of the drone is performed. This process simulates the impact of rain on various parts of the drone and calculates the local rain impact difference under different flight conditions. Due to changes in precipitation intensity and distribution, the impact of rain on different parts varies greatly. The calculation of the local rain impact difference can more accurately reflect the actual impact of rainfall on the structure and flight stability of the drone. This data provides a specific reference for the subsequent force imbalance structure analysis. Based on the local rain impact difference data, the flight state data and morphological structure data of the drone are decomposed into a force imbalance structure. This analysis process aims to identify the structural force imbalance caused by rainfall and help evaluate the impact of rain impact on the stability of the drone. Through a detailed analysis of the force imbalance, it can reveal the structural fatigue, deformation or instability of the drone under specific flight conditions and rainfall conditions. The decomposition data provides a basis for optimizing flight control strategies and structural design, ensuring that the drone can effectively respond to sudden changes in flight status under complex meteorological conditions.

[0021] Preferably, step S22 includes the following steps:

[0022] Step S221: extracting wind speed from the real-time wind data of the flight environment to obtain real-time wind speed change data;

[0023] Step S222: reconstructing the three-dimensional velocity field of the UAV flight state data according to the real-time velocity change data of the wind condition to obtain the three-dimensional velocity field of the flight state;

[0024] Step S223: calculating the fluid pressure difference at the morphological structure point of the UAV morphological structure data according to the three-dimensional velocity field of the flight state to obtain the fluid pressure difference at the morphological structure point;

[0025] Step S224: estimating the downward airflow pressure increase of the morphological contact surface based on the fluid pressure difference of the morphological structure point on the three-dimensional velocity field of the flight state, and obtaining downward airflow pressure increase data of the morphological contact surface;

[0026] Step S225: Perform structural fluid load evaluation based on the fluid pressure difference at the morphological structure point and the downward airflow pressurization data at the morphological contact surface to obtain structural fluid load data.

[0027] The present invention obtains real-time wind speed change data by extracting wind speed from real-time wind data in the flight environment. This process tracks the change trend and fluctuation characteristics of wind speed in real time by analyzing information such as wind speed and wind direction in sensor data. The change of wind speed directly affects the flight stability and aerodynamic characteristics of the UAV. Therefore, the accurate extraction of wind speed provides an important basis for the subsequent flight state evaluation, helps to analyze the influence of wind speed on the flight state and control system of the UAV, and ensures that the UAV can fly safely under different wind conditions. The three-dimensional velocity field of the flight state is obtained by reconstructing the three-dimensional velocity field of the UAV flight state data according to the real-time velocity change data of the wind condition. This reconstruction process combines the flight state data of the UAV (such as flight speed, attitude, etc.) and establishes an accurate three-dimensional velocity field model through numerical calculation methods. The velocity field model can describe the distribution of wind speed in three-dimensional space, help evaluate the airflow conditions encountered by the UAV under different flight states, and provide accurate data support for subsequent aerodynamic load calculation, structural analysis and attitude control. Based on the obtained three-dimensional velocity field data of the flight state, the fluid pressure difference of the morphological structure point is calculated for the morphological structure data of the UAV. This calculation simulates the interaction between wind flow and various parts of the drone and analyzes the pressure difference of airflow in different parts. By accurately calculating the fluid pressure difference of the drone structural points (such as wings, propellers, fuselage, etc.), we can further understand the distribution of aerodynamic loads during flight. This data provides a basis for subsequent fluid mechanics analysis and structural stability assessment, ensuring that the drone can maintain balance in a complex airflow environment. Based on the fluid pressure difference data of the morphological structure points, the morphological contact surface sinking airflow boosting is estimated for the three-dimensional velocity field of the flight state. This process predicts the boosting effect of airflow during the sinking of the contact surface by analyzing the interaction between the airflow and the contact surface of the drone structure. The boosting effect causes certain parts to bear greater airflow pressure, which in turn affects the stability and control performance of the drone. By accurately estimating this boosting effect, potential flight risks can be identified in advance, and data support can be provided for structural design and control system optimization to ensure the stability of the drone in different flight states. Combined with the fluid pressure difference of the morphological structure points and the sinking airflow boosting data of the morphological contact surface, the fluid load of the drone structure is evaluated. This process comprehensively considers the force of airflow on different parts of the drone, and obtains accurate structural fluid load data by calculating the influence of aerodynamic loads on the drone structure. These data can reveal the impact of wind speed, airflow pressure changes and pressurization effects on UAV flight performance, help optimize structural design and flight control strategies, and ensure that UAVs can maintain sufficient strength and stability in complex flight environments, thereby improving flight safety and reliability.

[0028] Preferably, step S24 comprises the following steps:

[0029] Step S241: performing rain impact density analysis between different regions of the UAV on the UAV morphological structure data according to the local rain impact difference data to obtain regional rain impact density data;

[0030] Step S242: performing mass center movement deviation trajectory identification on the UAV flight state data based on the local rain impact difference data and the regional rain impact density data to obtain mass center movement deviation trajectory data;

[0031] Step S243: evaluating the steering resistance limitation degree of the UAV flight state data according to the mass center motion deviation trajectory data to obtain steering resistance limitation degree data;

[0032] Step S244: quantifying the steering tremor of the steering resistance limitation degree data to obtain steering resistance tremor data;

[0033] Step S245: performing force imbalance structure decomposition on the UAV flight state data according to the mass center motion deviation trajectory data, the steering resistance limitation degree data and the steering resistance vibration data to obtain rain impact force imbalance decomposition data.

[0034] According to the local rain impact difference data, the present invention performs rain impact density analysis on different areas of the UAV to obtain regional rain impact density data. By analyzing the rain impact intensity of different areas during flight, it is possible to identify which parts are more significantly impacted. This analysis can help identify the difference in the impact of rain on the structure of the UAV, especially the impact on key areas (such as wings, tails, etc.), thereby providing guidance for structural optimization and protective measures, and improving the adaptability of the UAV under severe weather conditions. Based on the local rain impact difference data and the regional rain impact density data, the center of mass motion offset trajectory of the UAV flight state data is further identified to obtain the center of mass motion offset trajectory data. The density change of rain impact in different areas will cause the center of mass of the UAV to shift, and this shift affects the flight stability and control accuracy. By identifying the motion trajectory of the center of mass, the specific impact of rain impact on the flight attitude and trajectory of the UAV can be understood, which helps optimize the flight control strategy and adjust the flight path, thereby effectively reducing the instability in flight. According to the center of mass motion offset trajectory data, the steering resistance limitation degree of the UAV flight state is evaluated. This process aims to evaluate the resistance limitation encountered by the flight control system when turning under different flight states and rain impact conditions. Steering resistance refers to the additional torque and resistance that the drone needs to overcome when performing steering operations due to the effect of airflow or rain impact. By evaluating the degree of resistance limitation, the feasibility and efficiency of the drone's steering control when facing the influence of rain can be understood, which helps to adjust the flight control strategy to cope with these additional mechanical challenges. Steering tremor is quantified according to the steering resistance limitation degree data to obtain steering resistance tremor data. Steering tremor refers to the vibration or tremor phenomenon caused by uneven steering resistance or dynamic response of the structure when the drone performs steering operations. These tremors not only affect flight stability, but also increase the risk of fatigue damage to the structure. By quantifying steering tremor, the vibration problems encountered by the drone in different flight states and meteorological conditions can be accurately predicted, providing data support for the adjustment and structural design of the control system, and helping to reduce the impact of vibration on flight safety. According to the center of mass motion offset trajectory data, steering resistance limitation degree data and steering resistance tremor data, the flight state data of the drone is subjected to force imbalance structural decomposition to obtain the rain impact force imbalance decomposition data. This process integrates various data to identify the imbalance in flight status caused by rain impact, and then analyzes the force conditions in various areas of the drone. The force imbalance causes the drone to become unstable during flight, such as tilting, swaying or yaw. By decomposing these force imbalances, we can gain a deep understanding of the mechanical changes during flight and provide a basis for adjusting the flight control system, helping to optimize the flight performance and stability of the drone under complex weather conditions.

[0035] Preferably, step S243 includes the following steps:

[0036] According to the mass center motion offset trajectory data, the dynamic inertia moment of the UAV flight state data between different flight states is calculated to obtain the mass center offset dynamic inertia moment data set between different flight states;

[0037] Perform multi-interval steering angular velocity fluctuation fitting on the dynamic moment of inertia data set of mass center offset between different flight states to obtain steering angular velocity fluctuation fitting data;

[0038] According to the dynamic moment of inertia data set of mass center offset between different flight states and the steering angular velocity fluctuation fitting data, the transient steering pressure load between different flight states is calculated for the UAV flight state data to obtain the transient steering pressure load data;

[0039] The steering resistance limitation degree is evaluated based on the steering angular velocity fluctuation fitting data and the transient steering pressure load data to obtain the steering resistance limitation degree data.

[0040] The present invention calculates the dynamic moment of inertia of the flight state data of the unmanned aerial vehicle between different flight states according to the trajectory data of the center of mass motion deviation, and obtains a data set of dynamic moment of inertia of the center of mass deviation. This process calculates the dynamic moment of inertia under different flight states by analyzing the deviation of the center of mass caused by external forces (such as wind, rain impact, etc.) during flight, and combines the flight state data (such as speed, attitude, etc.). The change of the dynamic moment of inertia reflects the mass distribution and inertial characteristics of the unmanned aerial vehicle under different flight states, which is crucial to the stability of the flight control system. By establishing this data set, the aerodynamic load and mechanical response encountered during the flight process can be more accurately evaluated. The data set of the dynamic moment of inertia of the center of mass deviation between different flight states is fitted with multi-interval steering angular velocity fluctuations to obtain steering angular velocity fluctuation fitting data. By fitting and analyzing the steering angular velocity changes of the unmanned aerial vehicle under the flight state, the fluctuation characteristics of the steering angular velocity during flight can be identified. This fluctuation fitting process helps to reveal the dynamic response occurring in the steering operation, especially the instability phenomenon occurring during the steering process of the unmanned aerial vehicle under complex environments (such as the influence of factors such as strong wind and rain). The data can provide an important reference for the optimization of the flight control strategy, and avoid flight instability caused by excessive steering angular velocity fluctuations. According to the dynamic moment of inertia data set and the steering angular velocity fluctuation fitting data, the transient steering pressure load between different flight states of the UAV flight state data is calculated. The transient steering pressure load refers to how the aerodynamic pressure load of the UAV changes dynamically with the change of the steering angular velocity during the steering process. By combining the center of mass offset and the steering angular velocity fluctuation, the pressure load generated during the steering process can be calculated, which is crucial for evaluating the mechanical response of the UAV during the steering process. Accurate transient pressure load calculation helps ensure that the UAV can remain stable during rapid changes in direction or intense operations, and avoid structural damage or unstable control due to excessive pressure load. Based on the steering angular velocity fluctuation fitting data and the transient steering pressure load data, the steering resistance limitation degree of the UAV flight state is evaluated to obtain the steering resistance limitation degree data. The steering resistance limitation degree evaluation can help analyze the feasibility and efficiency of the UAV's steering operation under different flight states, and evaluate the responsiveness of its steering control system when facing external disturbances (such as wind speed and airflow changes). When the steering angular velocity fluctuates greatly, a large steering resistance will be generated, limiting the flexibility and maneuverability of the UAV. By evaluating the degree of steering resistance limitation, the flight control algorithm can be optimized, the control burden during steering can be reduced, and the flight safety and stability can be improved. This evaluation result provides a theoretical basis for dynamic control adjustments during flight missions.

[0041] Preferably, step S3 comprises the following steps:

[0042] Step S31: performing attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data;

[0043] Step S32: normalizing the flight state attitude control adjustment data to obtain flight state attitude control normalized data;

[0044] Step S32: Perform UAV control strategy logic learning based on the flight state attitude control normalization data to obtain the UAV control logic strategy.

[0045] The present invention adjusts the attitude control of the UAV flight state data by combining the rain impact force imbalance decomposition data and the structural fluid load data to obtain the flight state attitude control adjustment data. Rain impact and fluid load cause imbalance in the flight of the UAV and affect the flight stability. By analyzing the influence of force imbalance and fluid load, the attitude control of the UAV can be optimized and adjusted to compensate for the flight attitude deviation caused by external factors. Through this adjustment process, the balance control of the attitude of the UAV during flight can be achieved, thereby improving the accuracy and stability of the flight, especially the flight performance in bad weather. The flight state attitude control adjustment data is normalized to obtain the flight state attitude control normalized data. Normalization is to standardize the control data of different magnitudes and units so that they are in the same range or dimension, which is convenient for subsequent analysis and processing. This processing can eliminate the dimensional differences of different control parameters and ensure that each control data is effectively compared and processed under equal weight. Through the normalized data, the response efficiency of the control system can be improved, and standardized input is also provided for subsequent machine learning and control strategy optimization to ensure the stability and adaptability of the control strategy. Based on the normalized flight state attitude control data, the UAV control strategy logic is learned to obtain the UAV control logic strategy. By learning the normalized control data, the control system can automatically identify and optimize the control logic that best suits the current flight state. This process usually involves machine learning algorithms, such as reinforcement learning or deep learning, which continuously extract rules from flight data and adjust control parameters to improve the intelligence level of control decisions. The final control logic strategy will be able to automatically adjust the flight behavior of the UAV under different flight conditions, especially when dealing with uncertain factors such as environmental changes and external interference, to maintain flight stability and efficiency.

[0046] Preferably, step S31 includes the following steps:

[0047] Step S311: performing simulation calculation of abnormal numerical fluctuations of yaw angle and pitch angle of the UAV flight state data according to the unbalanced decomposition data of rain impact force and the structural fluid load data, and obtaining abnormal numerical fluctuation data of yaw angle and abnormal numerical fluctuation data of pitch angle respectively;

[0048] Step S312: performing a numerical instantaneous increment continuity analysis on the abnormal yaw angle numerical fluctuation data and the abnormal pitch angle numerical fluctuation data to obtain the abnormal instantaneous increment continuity data of the yaw angle and the abnormal instantaneous increment continuity data of the pitch angle;

[0049] Step S313: performing linkage coupling effect analysis based on the continuous data of the abnormal instantaneous increment of the yaw angle and the continuous data of the abnormal instantaneous increment of the pitch angle to obtain yaw / pitch coupling effect data;

[0050] Step S314: performing thrust vector correction on the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data according to the yaw / pitch coupling effect data to obtain flight thrust vector correction data;

[0051] Step S315: performing central inertia variable compensation correction on the flight thrust vector correction data to obtain central inertia variable compensation data;

[0052] Step S316: Perform attitude control adjustment on the UAV flight state data according to the flight thrust vector correction data and the central inertia variable compensation data to obtain flight state attitude control adjustment data.

[0053] According to the unbalanced force decomposition data of rain impact and the structural fluid load data, the present invention simulates and calculates the abnormal numerical fluctuations of the yaw angle and the pitch angle of the flight state of the unmanned aerial vehicle flight state data, and obtains the abnormal numerical fluctuation data of the yaw angle and the abnormal numerical fluctuation data of the pitch angle respectively. This process mainly simulates the abnormal fluctuations of the yaw and pitch angles caused by external disturbances (such as rain impact, airflow load, etc.) of the unmanned aerial vehicle, so as to identify the attitude deviation or instability occurring during the flight. By obtaining the fluctuation data of these two angles, the unbalanced state in the flight can be effectively predicted in advance, and data support and analysis basis can be provided for subsequent control adjustment, thereby improving the stability and reliability of the flight. The numerical instantaneous increment continuity analysis is performed on the abnormal numerical fluctuation data of the yaw angle and the abnormal numerical fluctuation data of the pitch angle, and the abnormal instantaneous increment continuity data of the yaw angle and the abnormal instantaneous increment continuity data of the pitch angle are obtained. By performing instantaneous increment analysis on these abnormal fluctuation data, the continuity of the yaw and pitch angle changes and their instantaneous increment characteristics can be captured during the flight, especially when subjected to external disturbances. The analysis can reveal the short-term drastic changes in the flight and provide early warning information for the attitude control system. The key significance of this step is to help the flight control system identify and respond to rapidly occurring control deviations, ensuring that the UAV can maintain stable attitude control during flight. Based on the continuous data of abnormal instantaneous increments of yaw angle and abnormal instantaneous increments of pitch angle, the linkage coupling effect analysis is performed to obtain the yaw / pitch coupling effect data. Abnormal changes in yaw and pitch angles often do not occur independently, and there is a coupling effect between them, especially when they are subject to external disturbances (such as wind force, airflow changes, etc.). This step analyzes the mutual influence of yaw and pitch angles to reveal how they are linked to change the attitude of the UAV during flight. By identifying this coupling effect, a more accurate control basis can be provided for the flight control system, so that the attitude can be adjusted more efficiently during flight, thereby improving the stability and maneuverability of the UAV in complex environments. According to the yaw / pitch coupling effect data, the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data are subjected to thrust vector correction to obtain the flight thrust vector correction data. Thrust vector correction is to compensate for the abnormal fluctuations of yaw and pitch angles caused by external disturbances (such as rain impact, wind speed changes, etc.) by adjusting the thrust direction of the UAV. This process ensures that when the aircraft encounters external disturbances, the thrust system can be adjusted in real time to reduce the impact of attitude changes. By correcting the thrust vector, the UAV can maintain a more stable flight attitude during flight, especially when facing sudden external climate factors, and can quickly restore the normal flight trajectory to ensure flight safety. The flight thrust vector correction data is corrected by the central inertia variable compensation to obtain the central inertia variable compensation data.The central inertia variable compensation correction compensates for the inertia change in the flight state to ensure that the center of gravity and moment of inertia of the aircraft are effectively controlled and corrected after the thrust vector correction. This step is mainly to calculate the inertial effect caused by the thrust direction adjustment of the drone during flight, and compensate the flight control according to these data to ensure that the control system of the aircraft will not have excessive control errors due to inertia changes. This process can improve the attitude control accuracy of the drone, so that the aircraft can maintain good flight performance and stability even in long-term flight or complex environments. According to the flight thrust vector correction data and the central inertia variable compensation data, the attitude control adjustment of the drone flight state data is performed to obtain the flight state attitude control adjustment data. By combining the thrust vector correction data and the inertia compensation data, the attitude control system of the aircraft can be accurately adjusted to ensure that it can still maintain a stable flight state in the face of external disturbances, internal load changes, etc. Attitude control adjustment is the core link in drone flight control, ensuring the precise control of the aircraft attitude during flight. Through this step, more efficient and stable flight control can be achieved, especially in complex environmental conditions, avoiding abnormal attitude of the aircraft, thereby improving the safety and stability of flight.

[0054] Preferably, step S314 includes the following steps:

[0055] According to the yaw / pitch coupling effect data, the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data are used to calculate the yaw and pitch flight resistance distribution difference, so as to obtain the yaw / pitch resistance distribution difference data;

[0056] Calculate the moment balance requirement for the yaw and pitch resistance distribution difference data to obtain the resistance difference moment balance requirement data;

[0057] Performing thrust vector direction correction processing between yaw angle / pitch angle on yaw and pitch resistance distribution difference data to obtain thrust vector direction correction data;

[0058] Based on the drag difference torque balance requirement data and the thrust vector direction correction data, the thrust vector distribution is optimized to obtain the thrust vector distribution optimization data;

[0059] The thrust vector correction is performed according to the drag difference torque balance requirement data, the thrust vector direction correction data and the thrust vector distribution optimization data to obtain the flight thrust vector correction data.

[0060] The present invention is based on the yaw / pitch coupling effect data, and the yaw angle numerical abnormal fluctuation data and the pitch angle numerical abnormal fluctuation data are calculated to obtain the yaw / pitch resistance distribution difference data. Flight resistance is an important factor affecting the stable flight of unmanned aerial vehicles. Especially in complex environments, the changes in yaw and pitch angles will affect the aerodynamic performance of the aircraft. By calculating the resistance distribution difference of yaw and pitch angles, the different effects of the two on the aircraft can be quantitatively revealed, which helps to identify the non-uniform air resistance distribution that occurs during flight. This calculation provides a key basis for subsequent control strategies and helps to design more accurate flight adjustment schemes. The torque balance requirement calculation is performed on the yaw and pitch resistance distribution difference data to obtain the resistance difference torque balance requirement data. During flight, due to the change of air resistance, the difference in yaw and pitch angles will cause torque imbalance, thereby affecting the attitude control of the aircraft. By calculating the torque balance requirement calculation of the resistance difference data, the torque amount that needs to be compensated can be determined. This process ensures that the torque changes caused by aerodynamic effects during flight can be corrected in a timely and effective manner, thereby ensuring that the aircraft maintains a stable attitude and reducing flight deviations caused by torque imbalance. The thrust vector direction correction processing between the yaw angle / pitch angle is performed on the yaw-pitch drag distribution difference data to obtain the thrust vector direction correction data. The direction correction of the thrust vector is a key step to ensure the stability of the aircraft attitude, especially when encountering abnormal yaw and pitch angles. By analyzing the drag distribution difference, the direction of the thrust vector can be adjusted so that the thrust is more matched with the attitude change of the aircraft. In this way, the adjustment of the thrust vector can effectively offset the imbalance caused by the fluctuation of the yaw and pitch angles, thereby ensuring that the control system of the aircraft can respond to external disturbances more accurately and improve the stability and safety of the flight. Based on the drag difference torque balance requirement data and the thrust vector direction correction data, the thrust vector distribution is optimized to obtain the thrust vector distribution optimization data. The goal of the thrust vector distribution optimization is to ensure that the aircraft can maintain a stable attitude under various flight conditions by reasonably configuring the output of each thrust unit. Combining the drag difference torque and thrust vector direction correction data, the thrust distribution can be optimized so that the thrust is reasonably proportioned in different parts of the aircraft. Through this optimization process, the aircraft attitude instability caused by uneven thrust distribution can be minimized, and the flight balance and control performance can be improved, especially in complex airflow or strong wind conditions. The optimized thrust distribution can better cope with external disturbances. According to the drag difference torque balance requirement data, thrust vector direction correction data, and thrust vector distribution optimization data, thrust vector correction is performed to obtain flight thrust vector correction data. Thrust vector correction is the final adjustment step of aircraft attitude control. Through comprehensive analysis of various data, the thrust vector correction can accurately compensate for the aircraft attitude deviation caused by drag difference and torque imbalance.The corrected thrust vector can more accurately adjust the flight trajectory of the aircraft, ensuring that the aircraft's attitude control is more stable and efficient. Through this correction process, the thrust distribution and direction can be optimized in real time when the external environment changes or the flight conditions are unstable, ensuring that the aircraft is always in the best flight state and improving flight safety and stability.

[0061] Preferably, the present invention further provides a drone control system for executing the drone control method as described above, the drone control system comprising:

[0062] The data acquisition module is used to collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, so as to obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, so as to obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively;

[0063] The force imbalance decomposition module is used to evaluate the structural fluid load of the UAV flight status data according to the real-time wind condition data of the flight environment to obtain the structural fluid load data; based on the structural fluid load data, the force imbalance structural decomposition of the flight environment rainfall characteristic data is performed to obtain the rain impact force imbalance decomposition data;

[0064] The attitude control adjustment module is used to perform attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain the flight state attitude control adjustment data; perform UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy;

[0065] The execution firmware design module is used to automatically execute the firmware design for the drone control logic strategy, obtain the drone control logic strategy firmware, and embed the drone control logic strategy firmware into the drone control center to execute the drone control method.

[0066] The beneficial effects of the present invention are that the present invention is an improvement on a traditional UAV control method, which solves the problems of inaccurate analysis of wind conditions and rainfall effects and large errors in UAV flight attitude control in the traditional UAV control method, improves the accuracy of analysis of wind conditions and rainfall effects, and reduces the errors in UAV flight attitude control. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of the steps of a method for controlling a drone;

[0068] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0069] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0070] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0071] See also Figures 1 to 3 , a drone control method, the method comprising the following steps:

[0072] Step S1: collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, and obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, and obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively;

[0073] Step S2: Perform structural fluid load evaluation on the UAV flight status data according to the real-time wind condition data of the flight environment to obtain structural fluid load data; perform force imbalance structural decomposition on the flight environment rainfall characteristic data based on the structural fluid load data to obtain rain impact force imbalance decomposition data;

[0074] Step S3: adjusting the attitude control of the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data; performing UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy;

[0075] Step S4: Automated execution firmware design is performed on the drone control logic strategy to obtain the drone control logic strategy firmware, and the drone control logic strategy firmware is embedded into the drone control center to execute the drone control method.

[0076] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a drone control method of the present invention. In this example, the drone control method includes the following steps:

[0077] Step S1: collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, and obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, and obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively;

[0078] In an embodiment of the present invention, the UAV collects environmental data and flight status data in real time during flight through the high-precision sensors (such as laser rangefinders, infrared rain gauges and anemometers) carried by the UAV. Environmental data include airflow velocity, wind direction, humidity and rainfall characteristics, and flight status data include the speed, heading, pitch angle, yaw angle and altitude change of the UAV. After data collection, the frequency domain characteristics of the wind condition data are extracted using a signal decomposition method based on Fourier transform, and the collected airflow data is parsed into instantaneous wind speed and wind direction distribution to form real-time wind condition data of the flight environment. At the same time, the raindrop impact frequency and intensity output by the rain gauge signal are analyzed by the histogram distribution method, and the rainfall characteristic parameters are extracted to generate the flight environment rainfall characteristic data.

[0079] Step S2: Perform structural fluid load evaluation on the UAV flight status data according to the real-time wind condition data of the flight environment to obtain structural fluid load data; perform force imbalance structural decomposition on the flight environment rainfall characteristic data based on the structural fluid load data to obtain rain impact force imbalance decomposition data;

[0080] In the embodiment of the present invention, based on the real-time wind data of the flight environment obtained in step S1, the fluid dynamics equation is established by using the Lagrangian method based on incompressible fluid dynamics, the surface fluid force characteristics of the UAV body are combined with the morphological structure data, and the transient load distribution of the UAV body under the action of the airflow is calculated point by point to obtain the structural fluid load data. Then, through the rainfall characteristic data, the impact force decomposition method is applied, and the force difference caused by the raindrop impact is calculated by partitioning with the raindrop kinetic energy, impact area and force time as parameters, and the local load under the raindrop impact is decomposed, and the imbalance torque of different parts is extracted, and finally the rain impact force imbalance decomposition data is output.

[0081] Step S3: adjusting the attitude control of the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data; performing UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy;

[0082] In an embodiment of the present invention, an attitude vector analysis method is used to perform real-time correction on the flight status data of the UAV based on the unbalanced decomposition data of the rain impact force and the structural fluid load data. First, the fluctuations of the yaw angle, pitch angle and roll angle are dynamically adjusted through attitude angle error calculation to ensure the stability of the attitude of the UAV. The high-order difference method is used to estimate the position of the center of inertia of the UAV and optimize the center of mass distribution. Next, the correction value of the thrust distribution of the UAV is calculated based on the principle of vector superposition to eliminate the attitude deviation. After the attitude adjustment is completed, the real-time adjusted flight status data is input into the control strategy optimization module based on the genetic algorithm, and the UAV control logic strategy is generated through multiple generations of loop iterations.

[0083] Step S4: Automated execution firmware design is performed on the drone control logic strategy to obtain the drone control logic strategy firmware, and the drone control logic strategy firmware is embedded into the drone control center to execute the drone control method.

[0084] In the embodiment of the present invention, according to the drone control logic strategy generated in step S3, the logic is coded using a hardware description language to realize the automated firmware design for a specific flight control unit. Through the logic mapping at the hardware circuit level, the drone control logic is converted into a high-speed execution firmware instruction set. After the firmware design is completed, the firmware is functionally tested using an integrated debugging platform to verify the reliability and execution efficiency of the logic. Finally, the firmware is written into the core processing unit of the drone control center through an embedded system program to realize the overall execution process of the drone control method.

[0085] Step S1 includes the following steps:

[0086] Step S11: collecting real-time data of the drone's flight environment and flight status through sensors carried by the drone, and obtaining real-time data of the drone's flight environment and flight status data;

[0087] Step S12: Fill missing values ​​in the real-time data of the UAV flight environment to obtain real-time filled data of the flight environment;

[0088] Step S13: extracting real-time wind condition data and rainfall characteristic data from the real-time filled data of the flight environment, and obtaining real-time wind condition data of the flight environment and rainfall characteristic data of the flight environment respectively.

[0089] In an embodiment of the present invention, the multi-type sensor modules carried by the UAV are used to collect flight environment data and flight status data in real time during the flight. The sensor module includes a high-precision wind speed and direction sensor, an optical rainfall sensor, a barometer, and an inertial measurement unit (IMU). The flight environment data includes environmental parameters such as wind speed, wind direction, rainfall, air pressure, temperature and humidity, and the flight status data includes dynamic information such as the speed, heading angle, pitch angle, roll angle, height change, and acceleration of the UAV. The data collection adopts a synchronous sampling strategy, and the sensor performs synchronous measurement at a fixed time interval (such as 50 milliseconds), and the data is transmitted to the UAV airborne data processing unit through a serial communication protocol for preliminary storage. During the collection process, in order to ensure data integrity, the cyclic redundancy check (CRC) technology is used to perform real-time verification of the sensor output data to eliminate the erroneous data generated during the transmission process. Finally, multi-dimensional flight environment real-time data and flight status data including timestamps, sensor type identifiers, and measurement values ​​are obtained. The real-time flight environment data collected in step S11 is preprocessed, and the data is firstly checked for integrity to identify missing values. Missing values ​​are caused by sensor failure, communication interruption or other interference factors. In the preprocessing process, a joint filling method based on time series interpolation and multidimensional correlation analysis is used to handle missing values. First, the missing values ​​in the continuous time period are filled in time series by linear interpolation, and the linear interpolation values ​​of the adjacent sampling points before and after the missing points are calculated. For non-continuous missing values ​​or isolated missing points, a weighted average method based on multidimensional features is used to infer the missing values ​​using the linear regression relationship between relevant parameters such as wind speed, humidity, and temperature. After filling processing, the real-time filling data of the flight environment is obtained, which meets the requirements of continuity and integrity. The wind condition and rainfall feature extraction is performed on the real-time filling data of the flight environment obtained in step S12. The wind condition feature extraction adopts a method based on vector decomposition and statistical analysis to solve the wind speed and wind direction components into distribution vectors, and calculate the change rate and direction fluctuation amplitude of the instantaneous wind speed to generate the real-time wind condition data of the flight environment. The specific operation includes: solving the wind speed and wind direction data based on the polar coordinate system, and calculating the distribution density of the wind vector. The wind speed change rate and fluctuation amplitude are extracted by the sliding window method, and the wind speed frequency distribution in the continuous time period is statistically analyzed. The wind speed sequence is spectrally analyzed using Fast Fourier Transform (FFT) to extract the main frequency components and turbulence characteristics of the airflow. Rainfall feature extraction is based on the output of the optical rain sensor, combined with the raindrop impact intensity signal, and the effective raindrop signal is separated by the dynamic threshold method. The extracted signal is statistically analyzed to extract characteristic parameters such as raindrop impact frequency, average particle size, and total rainfall intensity to form flight environment rainfall characteristic data. In this process, optical particle size analysis technology is used to estimate the distribution range of raindrops to ensure the accuracy of the characteristic data.

[0090] Step S2 includes the following steps:

[0091] Step S21: Acquire the morphological structure data of the drone;

[0092] Step S22: performing structural fluid load evaluation on the UAV flight state data and the UAV morphological structure data according to the real-time wind condition data of the flight environment to obtain structural fluid load data;

[0093] Step S23: Calculate the difference of local rain impact on the UAV based on the structural fluid load data and the rainfall characteristic data of the flight environment to obtain local rain impact difference data;

[0094] Step S24: performing force imbalance structural decomposition on the UAV flight state data and the UAV morphological structure data according to the local rain impact difference number to obtain rain impact force imbalance decomposition data.

[0095] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0096] Step S21: Acquire the morphological structure data of the drone;

[0097] In an embodiment of the present invention, the acquisition of the morphological structure data of the UAV is completed by accurately measuring the geometric dimensions, material properties and structural layout of the UAV. First, the laser scanning technology is used to perform three-dimensional morphological mapping of the entire UAV to generate point cloud data containing the shape contour and surface features. Secondly, based on the finite element modeling tool, the point cloud data is converted into a digital three-dimensional model, and the position, connection method and material properties of the key components of the UAV are marked, such as wing thickness, blade material and strength coefficient. In order to accurately capture the mechanical properties and density distribution of the material, a non-contact ultrasonic thickness gauge is used to measure the wall thickness of each part of the body, and a scanning electron microscope is used to analyze the microstructure of the material. Finally, the three-dimensional model and material properties are integrated into a unified data format to form the morphological structure data of the UAV, which provides basic parameters for subsequent fluid load calculations. Secondly, the morphological structure data of the UAV and related UAV manufacturing parameters can also be obtained through the UAV manufacturer.

[0098] Step S22: performing structural fluid load evaluation on the UAV flight state data and the UAV morphological structure data according to the real-time wind condition data of the flight environment to obtain structural fluid load data;

[0099] In an embodiment of the present invention, the structural fluid load evaluation is performed on the UAV flight status data and the UAV morphological structure data according to the real-time wind condition data of the flight environment. First, a vector description of the wind field is established using aerodynamic theory based on the wind speed, wind direction and turbulence characteristics in the real-time wind condition data of the flight environment. Then, the relative angle and windward area between each component of the UAV and the wind field are calculated using parameters such as the flight speed and attitude angle in the UAV flight status data. The Euler method of fluid mechanics is used to divide the three-dimensional model of the UAV into finite difference grids to simulate the pressure distribution of the wind field on the UAV structure. The local pressure changes when the airflow bypasses the UAV are calculated by vortex generation analysis to obtain the surface force values ​​of each component. Finally, the surface pressure distribution is accumulated through numerical integration to form the overall structural fluid load data.

[0100] Step S23: Calculate the difference of local rain impact on the UAV based on the structural fluid load data and the rainfall characteristic data of the flight environment to obtain local rain impact difference data;

[0101] In an embodiment of the present invention, based on the structural fluid load data and the flight environment rainfall characteristic data, the rain impact intensity of the local area of ​​the UAV surface is calculated, and the rain impact difference of each component is determined. First, the raindrop particle size distribution, impact frequency and velocity information in the rainfall characteristic data are used in combination with the fluid dynamics theory of the surface characteristics of the UAV to establish a mathematical model of the rain impact force. Then, the key position points on the surface of the UAV are selected, and the additional force changes on these position points during the raindrop impact process are calculated based on the pressure distribution values ​​provided by the structural fluid load data. The rain impact difference of each point is solved by the difference method, and the difference is archived in a table form to generate local rain impact difference data. This data provides differentiation parameters for the subsequent decomposition of the unbalanced force structure.

[0102] Step S24: performing force imbalance structural decomposition on the UAV flight state data and the UAV morphological structure data according to the local rain impact difference number to obtain rain impact force imbalance decomposition data.

[0103] In the embodiment of the present invention, the flight state data and the morphological structure data of the UAV are decomposed into a force imbalance structure according to the local rain impact difference data, and the mechanical imbalance state of each structural area caused by rain impact is analyzed. First, according to the change of the additional force value of each component recorded in the rain impact difference data, the morphological structure of the UAV is decomposed into several force units, and the coupling effect between the units is determined according to the structural topological relationship.

[0104] Then, the force difference of the force unit in each dimension is calculated by using the equilibrium moment theory, and the regional force imbalance distribution map is generated through matrix operation. Finally, the data in the force imbalance distribution map is converted into a linear decomposition matrix to clarify the dynamic deformation trend of the UAV in the local area due to rain impact, and the decomposition data of the force imbalance caused by rain impact is obtained. This data can provide a scientific basis for the design of the UAV attitude adjustment strategy.

[0105] Step S22 includes the following steps:

[0106] Step S221: extracting wind speed from the real-time wind data of the flight environment to obtain real-time wind speed change data;

[0107] Step S222: reconstructing the three-dimensional velocity field of the UAV flight state data according to the real-time velocity change data of the wind condition to obtain the three-dimensional velocity field of the flight state;

[0108] Step S223: calculating the fluid pressure difference at the morphological structure point of the UAV morphological structure data according to the three-dimensional velocity field of the flight state to obtain the fluid pressure difference at the morphological structure point;

[0109] Step S224: estimating the downward airflow pressure increase of the morphological contact surface based on the fluid pressure difference of the morphological structure point on the three-dimensional velocity field of the flight state, and obtaining downward airflow pressure increase data of the morphological contact surface;

[0110] Step S225: Perform structural fluid load evaluation based on the fluid pressure difference at the morphological structure point and the downward airflow pressurization data at the morphological contact surface to obtain structural fluid load data.

[0111] In the embodiment of the present invention, the wind speed extraction is completed by vector decomposition and statistical analysis of the real-time wind data of the flight environment. First, the instantaneous value of the wind speed is extracted using the continuous time series of wind data collection, and a two-dimensional wind speed vector is constructed according to the wind speed direction at each time point. The Fourier transform method is used to filter out the high-frequency noise in the time series, and only the main wind speed change trend is retained. Then, the deviation caused by sensor noise or interference in the measurement data is dynamically corrected by the Kalman filter algorithm to generate a time change curve of the real-time wind speed. The rate of change of wind speed in a continuous time period is calculated by the difference method, and a three-dimensional spatial description of wind speed change is formed in combination with wind direction data to obtain real-time wind speed change data. This data provides basic input parameters for subsequent speed field reconstruction. The reconstruction of the three-dimensional speed field of the flight state is completed based on the real-time speed change data of the wind condition and the flight speed, attitude angle and other parameters recorded in the flight state data of the drone. First, the wind speed change data is aligned with the flight path coordinates by linear interpolation to ensure the consistency of the two sets of data in space and time. Secondly, the flight speed of the UAV is decomposed into three-dimensional components, and the relative velocity field of the UAV in the flight environment is calculated by vector superposition. Combined with the attitude angle and tilt angle data of the UAV, the rotation matrix is ​​used to transform the coordinates of the velocity field to ensure that the velocity field can accurately describe the actual wind state of each component of the UAV. Finally, the velocity field description of the three-dimensional space around the UAV is generated by the difference algorithm to form the three-dimensional velocity field data of the flight state. The calculation of the fluid pressure difference of the morphological structure point is based on the three-dimensional velocity field of the flight state and is completed in combination with the geometric point set in the UAV morphological structure data. First, according to the magnitude and direction of each velocity vector in the velocity field, the Bernoulli equation is used to calculate the static pressure and dynamic pressure of each structural point on the surface of the UAV. Subsequently, the finite difference method is used to calculate the difference between the pressure values ​​of adjacent points to determine the fluid pressure gradient of each local area on the surface of the UAV. Through the pressure difference analysis of the grid distribution, a matrix data describing the fluid pressure distribution on the UAV morphological surface is generated. The matrix accurately reflects the difference in fluid pressure on each component of the UAV during flight and is recorded as the fluid pressure difference data of the morphological structure point. The estimation of the downdraft pressure of the morphological contact surface is completed based on the fluid pressure difference of the morphological structure point and the three-dimensional velocity field data of the flight state. First, by analyzing the fluid pressure difference at the morphological structure point, the local sinking area on the UAV surface caused by the pressure difference is determined. The pressure difference is fitted with the vertical component of the velocity field using the linear regression method to quantify the airflow pressurization effect in the sinking area. Secondly, based on the airflow distribution described in the three-dimensional velocity field of the flight state, the continuity equation of fluid dynamics is used to simulate the acceleration change of the airflow in the sinking area, and the airflow pressurization coefficient is calculated. Finally, the airflow pressurization coefficient is combined with the geometric dimensions of the sinking area to generate the sinking airflow pressurization data of the morphological contact surface, which is used to describe the airflow mechanics changes caused by the sinking effect of each component. The structural fluid load assessment is completed by integrating the fluid pressure difference at the morphological structure point and the sinking airflow pressurization data of the morphological contact surface.First, based on the position of the UAV's morphological structure points, the fluid pressure difference matrix is ​​mapped to the UAV's three-dimensional morphological model to form a force map of the structural surface. Then, the additional effect of local airflow pressurization on the overall structural load of the UAV is calculated using the downward airflow pressurization data of the morphological contact surface. Through the numerical integration method, the pressure difference and airflow pressurization effects are added to the overall force analysis of the UAV structural model to determine the total fluid load of the UAV in flight. Finally, the integrated structural fluid load data is output to provide data support for the subsequent UAV flight control strategy design.

[0112] Step S24 includes the following steps:

[0113] Step S241: performing rain impact density analysis between different regions of the UAV on the UAV morphological structure data according to the local rain impact difference data to obtain regional rain impact density data;

[0114] Step S242: performing mass center movement deviation trajectory identification on the UAV flight state data based on the local rain impact difference data and the regional rain impact density data to obtain mass center movement deviation trajectory data;

[0115] Step S243: evaluating the steering resistance limitation degree of the UAV flight state data according to the mass center motion deviation trajectory data to obtain steering resistance limitation degree data;

[0116] Step S244: quantifying the steering tremor of the steering resistance limitation degree data to obtain steering resistance tremor data;

[0117] Step S245: performing force imbalance structure decomposition on the UAV flight state data according to the mass center motion deviation trajectory data, the steering resistance limitation degree data and the steering resistance vibration data to obtain rain impact force imbalance decomposition data.

[0118] In the embodiment of the present invention, the rain impact density analysis is performed based on the local rain impact difference data and the UAV morphological structure data. First, the local rain impact difference data is mapped to the surface grid point set of the UAV morphological structure data to determine the rain impact intensity of each grid point. The raindrop impact frequency is fitted by Poisson distribution, and the rain impact energy per unit area is calculated by the integral method and quantified as the rain impact density value. Secondly, the surface area is divided into several sub-areas by using the geometric characteristics of the UAV morphology, and the regional rain impact density data is generated according to the impact density distribution of the grid points in each sub-area. The impact density is evaluated by statistical analysis to provide an accurate impact density distribution map for subsequent analysis. The center of mass motion offset trajectory recognition is based on the local rain impact difference data and the regional rain impact density data, and is completed by analyzing the dynamic characteristics of the UAV flight state data. First, the total torque of the UAV under the action of rain impact is calculated to evaluate the dynamic offset trend of its center of mass. The particle dynamics formula is used to combine the rain impact density and the UAV attitude angle data to calculate the trajectory offset of the center of mass in three-dimensional space. The trajectory deviation curve is smoothed by interpolation method to remove measurement noise interference. Finally, the position of the key turning point in the deviation trajectory is identified by the gradient descent algorithm, and the center of mass motion deviation trajectory data is generated for subsequent steering resistance evaluation. The steering resistance limitation degree evaluation is completed based on the center of mass motion deviation trajectory data and the flight angular velocity and attitude changes recorded in the UAV flight status data. First, the magnitude of the drag torque caused by rain impact is calculated by combining the deviation direction and amplitude of the center of mass trajectory. Subsequently, the attitude angular velocity changes in the flight status data are analyzed to evaluate the response sensitivity of the UAV under different flight conditions. The extreme value analysis method is used to extract the maximum resistance point of the steering action during flight and quantify the steering limitation degree. Comprehensive steering resistance limitation degree data is generated by the cumulative integral method to describe the impact of rain impact on steering ability. The steering resistance tremor quantification is based on the steering resistance limitation degree data and combined with the UAV flight status data for multi-spectral analysis. First, the frequency components in the steering resistance limitation degree data are decomposed by fast Fourier transform (FFT) to identify high-frequency tremor signals. Then, the tremor signal is jointly analyzed in the time and frequency domains by wavelet transform to locate the occurrence area and intensity change of the tremor. Combined with the corresponding rigid parameters in the UAV morphological structure data, the numerical simulation method is used to calculate the dynamic interference degree of different frequency vibrations on the UAV's steering ability, generate steering resistance vibration data, and provide input for subsequent structural decomposition. The force imbalance structural decomposition is completed through mechanical balance analysis with the center of mass motion offset trajectory data, steering resistance limitation degree data and steering resistance vibration data as input. First, the above data are fused and mapped to the three-dimensional model of the UAV morphological structure to determine the force imbalance characteristics of the local area.Then, the finite element analysis method is used to simulate the dynamic mechanical response of the UAV under the impact of rain, and the key nodes and paths of force imbalance in the structure are identified. Finally, the complex force imbalance area is divided into several independent sub-problems through the step-by-step decomposition method, and the decomposition data of the force imbalance under rain impact is generated to accurately describe the changes in the mechanical characteristics of the overall and local structures of the UAV, providing a basis for flight control adjustment.

[0119] Step S243 includes the following steps:

[0120] According to the mass center motion offset trajectory data, the dynamic inertia moment of the UAV flight state data between different flight states is calculated to obtain the mass center offset dynamic inertia moment data set between different flight states;

[0121] Perform multi-interval steering angular velocity fluctuation fitting on the dynamic moment of inertia data set of mass center offset between different flight states to obtain steering angular velocity fluctuation fitting data;

[0122] According to the dynamic moment of inertia data set of mass center offset between different flight states and the steering angular velocity fluctuation fitting data, the transient steering pressure load between different flight states is calculated for the UAV flight state data to obtain the transient steering pressure load data;

[0123] The steering resistance limitation degree is evaluated based on the steering angular velocity fluctuation fitting data and the transient steering pressure load data to obtain the steering resistance limitation degree data.

[0124] In the embodiment of the present invention, the dynamic moment of inertia calculation is completed based on the center of mass motion offset trajectory data and the UAV flight state data. First, the center of mass position change value and angular velocity change value of the UAV under different flight states are extracted. Using the rigid body dynamics formula, the center of mass offset is combined with the angular velocity to calculate the instantaneous moment of inertia of the UAV under each flight state. The specific operation includes using the particle dynamics method to obtain the movement amplitude and direction change of the center of mass by integrating the acceleration change of the center of mass trajectory curve. At the same time, using the rotational dynamics formula, the center of mass offset is combined with the rotation radius of the entire UAV to solve the change amplitude of the local moment of inertia. Finally, the calculation results are segmented in the time dimension to generate a dynamic moment of inertia data set of the center of mass offset between different flight states. This data set comprehensively describes the inertial dynamic characteristics of the UAV under the influence of rain impact and wind conditions, and provides a basis for subsequent analysis. The multi-interval steering angular velocity fluctuation fitting is completed with the dynamic moment of inertia data set as input and combined with the steering angular velocity record in the UAV flight state data. First, the angular velocity data recorded by the UAV during flight is segmented by time, and the flight state switching point is marked. The segmented linear regression method is used to fit the trend of each angular velocity change to generate a preliminary fluctuation fitting curve. Then, the nonlinear fluctuation in the steering angular velocity data is corrected by introducing a high-order polynomial fitting method, and the fitting curve is dynamically adjusted in combination with the change value of the moment of inertia in the dynamic moment of inertia data set. Finally, the fluctuation amplitude, frequency characteristics and change trend of the fitting curve are converted into steering angular velocity fluctuation fitting data, which are used to describe the dynamic change characteristics of the angular velocity during the steering process. The transient steering pressure load calculation is completed based on the dynamic moment of inertia data set and the steering angular velocity fluctuation fitting data. First, the angular velocity change rate in the steering angular velocity fluctuation fitting data is used to combine the flight speed and attitude angle parameters in the UAV flight state data to establish a mathematical model of the pressure load during the steering process. Subsequently, the Euler method is used to discretize the pressure load model, and the steering pressure load of the UAV at each time point is calculated by stepwise integration. The moment of inertia value in the dynamic moment of inertia data set is used as a weighting coefficient to correct the transient change of the pressure load in the local area. Finally, the transient steering pressure load data is generated to describe the dynamic pressure changes of the UAV during the steering process under different flight states. The steering resistance limitation degree assessment is completed through mechanical analysis with the steering angular velocity fluctuation fitting data and transient steering pressure load data as input. First, the transient steering pressure load data is mapped to the three-dimensional structural model of the UAV to identify the areas where each component is subjected to the greatest force during the steering process. Then, combined with the frequency characteristics in the steering angular velocity fluctuation fitting data, the damping mechanics theory is used to calculate the resistance change trend of the UAV in the force-bearing area. Through the multi-objective optimization algorithm, the extreme points of the steering resistance are extracted, and their influence on the steering flexibility is calculated.Finally, all analysis results are integrated into steering resistance limitation data to quantify the limiting effects of rain impact and flight environment on the steering performance of the UAV, providing an accurate basis for further attitude adjustment and control strategies.

[0125] Step S3 includes the following steps:

[0126] Step S31: performing attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data;

[0127] Step S32: normalizing the flight state attitude control adjustment data to obtain flight state attitude control normalized data;

[0128] Step S32: Perform UAV control strategy logic learning based on the flight state attitude control normalization data to obtain the UAV control logic strategy.

[0129] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0130] Step S31: performing attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data;

[0131] In an embodiment of the present invention, the process of adjusting the attitude control of the UAV flight state data according to the unbalanced force decomposition data of rain impact and the structural fluid load data includes the following specific operations. First, the unbalanced force decomposition data of rain impact and the structural fluid load data are vector-superimposed to calculate the synthetic force field of the UAV at different time points, and generate a dynamic force spectrum related to the flight state. Using the Euler angle transformation, the attitude change value of the UAV at each moment is matched with the dynamic force spectrum to identify the attitude deviation area caused by the force imbalance in the flight state. Then, the proportional-integral-differential (PID) control method is adopted to generate the corresponding attitude adjustment control instruction based on the real-time force change and the flight state deviation angle. In the specific implementation process, the adjustment instruction is decomposed into the attitude adjustment amount of the three axes of roll, pitch and yaw, and it is smoothed by numerical integration to ensure the stability of the adjustment process. Finally, the adjusted flight state record is used as the flight state attitude control adjustment data to provide a further attitude optimization basis.

[0132] Step S32: normalizing the flight state attitude control adjustment data to obtain flight state attitude control normalized data;

[0133] In an embodiment of the present invention, the process of normalizing the flight state attitude control adjustment data is centered on the numerical normalization method. First, the attitude change values ​​of the three dimensions of roll, pitch and yaw in the flight state attitude control adjustment data, as well as the adjustment amplitude, frequency and response time data of each axis are extracted. According to statistical analysis, the maximum and minimum values ​​of the data of each dimension are determined to calculate the standardized range. Then, using the linear transformation formula, the data of each dimension is mapped to the interval range of [0,1] to generate a preliminary normalized result. On this basis, the normalized data is secondary processed by the Gaussian smoothing algorithm to eliminate local discontinuities caused by sampling errors. Finally, the normalized data is saved as the normalized data of the flight state attitude control, which is used as the input of the logic learning of the drone control strategy.

[0134] Step S32: Perform UAV control strategy logic learning based on the flight state attitude control normalization data to obtain the UAV control logic strategy.

[0135] In an embodiment of the present invention, the process of performing logic learning of the UAV control strategy according to the normalized data of the flight state attitude control is completed based on the heuristic logic rule construction. First, the normalized data is segmented according to the time series, and the attitude adjustment mode in each time period is extracted, including the adjustment direction, amplitude and response rate. The correspondence table between the flight state and the control strategy is constructed by using the dynamic change trend of the normalized data. Subsequently, a rule generation algorithm is used to recursively search the feature points in the correspondence table to extract the optimal control strategy logic path. In the specific implementation process, the control logic of different time periods is cross-validated through the path backtracking method to ensure the robustness of the strategy generation. Finally, the generated logic rules are integrated into the UAV control logic strategy and saved as a standardized logic expression to guide subsequent flight control operations and optimization.

[0136] Step S31 includes the following steps:

[0137] Step S311: performing simulation calculation of abnormal numerical fluctuations of yaw angle and pitch angle of the UAV flight state data according to the unbalanced decomposition data of rain impact force and the structural fluid load data, and obtaining abnormal numerical fluctuation data of yaw angle and abnormal numerical fluctuation data of pitch angle respectively;

[0138] Step S312: performing a numerical instantaneous increment continuity analysis on the abnormal yaw angle numerical fluctuation data and the abnormal pitch angle numerical fluctuation data to obtain the abnormal instantaneous increment continuity data of the yaw angle and the abnormal instantaneous increment continuity data of the pitch angle;

[0139] Step S313: performing linkage coupling effect analysis based on the continuous data of the abnormal instantaneous increment of the yaw angle and the continuous data of the abnormal instantaneous increment of the pitch angle to obtain yaw / pitch coupling effect data;

[0140] Step S314: performing thrust vector correction on the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data according to the yaw / pitch coupling effect data to obtain flight thrust vector correction data;

[0141] Step S315: performing central inertia variable compensation correction on the flight thrust vector correction data to obtain central inertia variable compensation data;

[0142] Step S316: Perform attitude control adjustment on the UAV flight state data according to the flight thrust vector correction data and the central inertia variable compensation data to obtain flight state attitude control adjustment data.

[0143] In an embodiment of the present invention, the yaw angle and pitch angle in the flight state data of the unmanned aerial vehicle are simulated and calculated for abnormal fluctuations according to the unbalanced force decomposition data of rain impact and the structural fluid load data. First, the regional force distribution characteristics contained in the unbalanced force decomposition data of rain impact are extracted, and the dynamic force distribution diagram of the flight state is constructed in combination with the local pressure change value in the structural fluid load data. The response curves of the yaw angle and the pitch angle when the force changes suddenly are identified by matching the force distribution diagram with the time series of the flight state data. On this basis, the main frequency components of the yaw angle and the pitch angle changes are extracted by the Fourier transform method, and a frequency domain characteristic model of numerical abnormal fluctuations is constructed. By gradually increasing the pressure disturbance value to simulate the comprehensive influence of rain impact and fluid load on the flight state, the numerical abnormal fluctuation data of the yaw angle and the numerical abnormal fluctuation data of the pitch angle are obtained. The instantaneous increment persistence analysis is performed on the numerical abnormal fluctuation data of the yaw angle and the numerical abnormal fluctuation data of the pitch angle. First, the above abnormal fluctuation data is segmented according to the time series, and the instantaneous increment change amplitude and change speed in each segment are extracted. The instantaneous increment value in each segment of data is calculated by the differential method, and the noise is removed by the Gaussian filtering algorithm to ensure the stability of the calculation results. Subsequently, the instantaneous increment value is accumulated and integrated to generate a continuous change trend curve, and the correlation analysis is performed with the rain impact and fluid load change characteristics. Through the dynamic fitting of the trend line slope, the continuous data of the abnormal instantaneous increment of the yaw angle and the abnormal instantaneous increment of the pitch angle are obtained for the subsequent linkage effect analysis. Based on the continuous data of the abnormal instantaneous increment of the yaw angle and the abnormal instantaneous increment of the pitch angle, the linkage coupling effect analysis is carried out. First, the time point synchronization features in the above data are extracted to construct the joint change matrix of the yaw angle and the pitch angle. The singular value decomposition (SVD) method is used to decompose the matrix and extract its main coupling mode. By analyzing the eigenvectors of the main coupling mode, the coupling strength between the yaw angle and the pitch angle under different flight states is calculated, and the peak time point and response area of ​​the coupling effect are further identified. Finally, the yaw / pitch coupling effect data is output to provide a reference for the subsequent vector correction. According to the yaw / pitch coupling effect data, the thrust vector correction is performed on the abnormal fluctuation data of the yaw angle value and the abnormal fluctuation data of the pitch angle value. First, the UAV thrust distribution model is established to map the numerical changes of the abnormal fluctuations of the yaw and pitch angles to the adjustment direction and amplitude of the thrust vector. The corrected thrust distribution is calculated using the vector superposition method. To ensure the smoothness of the correction result, the Bezier curve interpolation method is used to iteratively optimize the thrust vector multiple times to gradually reduce the attitude deviation caused by rain impact and fluid load. The corrected results are stored as flight thrust vector correction data for further inertial variable compensation. The center inertial variable compensation is performed on the flight thrust vector correction data. First, the inertial center position and change trajectory of the UAV are calculated based on the corrected thrust vector.The real-time change of the center inertia of the UAV is calculated by analyzing the dynamic formula, and the high-frequency fluctuation components are extracted. The Kalman filter is used to dynamically correct the inertial variable to remove the influence of environmental interference and system noise, so as to obtain accurate inertial variable compensation results. Finally, the center inertial variable compensation data is generated to provide accurate basic input for attitude control adjustment. According to the flight thrust vector correction data and the center inertial variable compensation data, the attitude control adjustment of the UAV flight state data is performed. First, the attitude adjustment direction and adjustment amount of the UAV at each flight time point are calculated by combining the two types of input data. Using the Euler angle conversion formula, the adjustment amount is decomposed into specific values ​​in the three directions of roll, pitch and yaw. The proportional-integral-differential (PID) control strategy is adopted to input the attitude adjustment signal into the attitude control system of the UAV, and the control parameters are adjusted through the real-time feedback mechanism to ensure the dynamic stability of the flight state. Finally, the flight state attitude control adjustment data is output to provide reliable protection for the flight safety of the UAV.

[0144] Step S314 includes the following steps:

[0145] According to the yaw / pitch coupling effect data, the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data are used to calculate the yaw and pitch flight resistance distribution difference, so as to obtain the yaw / pitch resistance distribution difference data;

[0146] Calculate the moment balance requirement for the yaw and pitch resistance distribution difference data to obtain the resistance difference moment balance requirement data;

[0147] Performing thrust vector direction correction processing between yaw angle / pitch angle on yaw and pitch resistance distribution difference data to obtain thrust vector direction correction data;

[0148] Based on the drag difference torque balance requirement data and the thrust vector direction correction data, the thrust vector distribution is optimized to obtain the thrust vector distribution optimization data;

[0149] The thrust vector correction is performed according to the drag difference torque balance requirement data, the thrust vector direction correction data and the thrust vector distribution optimization data to obtain the flight thrust vector correction data.

[0150] In an embodiment of the present invention, according to the yaw / pitch coupling effect data, the flight resistance distribution difference calculation is performed on the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data. First, the force distribution characteristics in the yaw and pitch angle value abnormal fluctuation data are extracted, and the force change curves of the two are segmented to generate their respective resistance change matrices. The resistance change at different time points is extracted by matrix difference calculation. Subsequently, the resistance change is multiplied by the correlation coefficient in the yaw / pitch coupling effect data to obtain the resistance difference distribution value of the two in each flight state, and the cumulative distribution difference is calculated by the integral method. Finally, the yaw / pitch resistance distribution difference data is output to provide input for the subsequent moment balance calculation. The moment balance demand calculation is performed on the yaw pitch resistance distribution difference data. First, a moment distribution model of the UAV in three-dimensional space is established. The model takes the center of mass of the UAV as the reference and calculates the moment change caused by the resistance difference on the yaw and pitch angles. The instantaneous moment values ​​in the yaw and pitch directions are calculated point by point using the lever arm formula. The above calculation results are input into the torque balance equation group, and the additional thrust distribution required to achieve torque balance is solved through matrix inverse operation to generate the drag difference torque balance demand data. The calculation process of torque balance needs to ensure numerical continuity and stability to avoid jump fluctuations. The thrust vector direction between the yaw angle and the pitch angle is corrected for the yaw and pitch drag distribution difference data. According to the spatial characteristics of the drag distribution difference, the main force directions in the yaw and pitch directions are extracted, and the thrust vector direction correction formula is established. The decomposition result of the vector is superimposed with the yaw and pitch coupling effect data to recalculate the adjustment angle of the thrust vector. Subsequently, the corrected vector direction is calculated using the three-dimensional rotation matrix method (based on the Euler angle conversion formula), and it is verified whether the effect of the vector direction adjustment on the flight attitude meets the established force balance requirements. Finally, the thrust vector direction correction data is output to provide input for subsequent distribution optimization. Thrust vector distribution optimization is performed based on the drag difference torque balance demand data and the thrust vector direction correction data. First, the adjustment results in the thrust vector direction correction data are superimposed with the torque compensation value in the drag difference torque balance demand data to generate a preliminary thrust vector distribution map. Through a multi-objective optimization algorithm (such as a particle swarm optimization algorithm), the optimal solution for the thrust vector distribution is found while maintaining flight stability. Constraints are set during the optimization process to ensure that the adjustment range of the thrust vector does not exceed the structural bearing capacity of the drone. After the optimization is completed, the thrust vector distribution optimization data is generated to provide input for the final thrust vector correction. Thrust vector correction is performed based on the drag difference torque balance demand data, thrust vector direction correction data, and thrust vector distribution optimization data. The thrust vector correction formula is constructed by integrating the input of the three types of data, and the torque balance demand and the vector direction optimization results are integrated to calculate the final thrust distribution parameters.Through iterative correction, the numerical stability and accuracy of the thrust vector correction are ensured, and the output value of each thrust point of the drone is gradually adjusted. Finally, the flight thrust vector correction data is generated for the attitude control adjustment of the drone flight state to ensure the smooth operation of the drone in complex environments.

[0151] The present invention also provides a drone control system for executing the drone control method as described above, the drone control system comprising:

[0152] The data acquisition module is used to collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, so as to obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, so as to obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively;

[0153] The force imbalance decomposition module is used to evaluate the structural fluid load of the UAV flight status data according to the real-time wind condition data of the flight environment to obtain the structural fluid load data; based on the structural fluid load data, the force imbalance structural decomposition of the flight environment rainfall characteristic data is performed to obtain the rain impact force imbalance decomposition data;

[0154] The attitude control adjustment module is used to perform attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain the flight state attitude control adjustment data; perform UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy;

[0155] The execution firmware design module is used to automatically execute the firmware design for the drone control logic strategy, obtain the drone control logic strategy firmware, and embed the drone control logic strategy firmware into the drone control center to execute the drone control method.

[0156] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A drone control method, characterized in that: The following steps are involved: Step S1: collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, and obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, and obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively; Step S2: performing structural fluid load evaluation on the UAV flight status data according to the real-time wind condition data of the flight environment to obtain structural fluid load data; Based on the structural fluid load data, the flight environment rainfall characteristic data is decomposed into a force imbalance structure to obtain the rain impact force imbalance decomposition data; Step S3: adjusting the attitude control of the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data; performing UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy; Step S4: Automated execution firmware design is performed on the drone control logic strategy to obtain the drone control logic strategy firmware, and the drone control logic strategy firmware is embedded into the drone control center to execute the drone control method.

2. The drone control method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: collecting real-time data of the drone's flight environment and flight status through sensors carried by the drone, and obtaining real-time data of the drone's flight environment and flight status data; Step S12: Fill missing values ​​in the real-time data of the UAV flight environment to obtain real-time filled data of the flight environment; Step S13: extracting real-time wind condition data and rainfall characteristic data from the real-time filled data of the flight environment, and obtaining real-time wind condition data of the flight environment and rainfall characteristic data of the flight environment respectively.

3. The drone control method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: Acquire the morphological structure data of the drone; Step S22: performing structural fluid load evaluation on the UAV flight state data and the UAV morphological structure data according to the real-time wind condition data of the flight environment to obtain structural fluid load data; Step S23: Calculate the difference of local rain impact on the UAV based on the structural fluid load data and the rainfall characteristic data of the flight environment to obtain local rain impact difference data; Step S24: performing force imbalance structural decomposition on the UAV flight state data and the UAV morphological structure data according to the local rain impact difference number to obtain rain impact force imbalance decomposition data.

4. The drone control method according to claim 3, characterized in that: Step S22 includes the following steps: Step S221: extracting wind speed from the real-time wind data of the flight environment to obtain real-time wind speed change data; Step S222: reconstructing the three-dimensional velocity field of the UAV flight state data according to the real-time velocity change data of the wind condition to obtain the three-dimensional velocity field of the flight state; Step S223: calculating the fluid pressure difference at the morphological structure point of the UAV morphological structure data according to the three-dimensional velocity field of the flight state to obtain the fluid pressure difference at the morphological structure point; Step S224: estimating the downward airflow pressure increase of the morphological contact surface based on the fluid pressure difference of the morphological structure point on the three-dimensional velocity field of the flight state, and obtaining downward airflow pressure increase data of the morphological contact surface; Step S225: Perform structural fluid load evaluation based on the fluid pressure difference at the morphological structure point and the downward airflow pressurization data at the morphological contact surface to obtain structural fluid load data.

5. The drone control method according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: performing rain impact density analysis between different regions of the UAV on the UAV morphological structure data according to the local rain impact difference data to obtain regional rain impact density data; Step S242: performing mass center movement deviation trajectory identification on the UAV flight state data based on the local rain impact difference data and the regional rain impact density data to obtain mass center movement deviation trajectory data; Step S243: evaluating the steering resistance limitation degree of the UAV flight state data according to the mass center motion deviation trajectory data to obtain steering resistance limitation degree data; Step S244: quantifying the steering tremor of the steering resistance limitation degree data to obtain steering resistance tremor data; Step S245: performing force imbalance structure decomposition on the UAV flight state data according to the mass center motion deviation trajectory data, the steering resistance limitation degree data and the steering resistance vibration data to obtain rain impact force imbalance decomposition data.

6. The drone control method according to claim 5, characterized in that: Step S243 includes the following steps: According to the mass center motion offset trajectory data, the dynamic inertia moment of the UAV flight state data between different flight states is calculated to obtain the mass center offset dynamic inertia moment data set between different flight states; Perform multi-interval steering angular velocity fluctuation fitting on the dynamic moment of inertia data set of mass center offset between different flight states to obtain steering angular velocity fluctuation fitting data; According to the dynamic moment of inertia data set of mass center offset between different flight states and the steering angular velocity fluctuation fitting data, the transient steering pressure load between different flight states is calculated for the UAV flight state data to obtain the transient steering pressure load data; The steering resistance limitation degree is evaluated based on the steering angular velocity fluctuation fitting data and the transient steering pressure load data to obtain the steering resistance limitation degree data.

7. The drone control method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain flight state attitude control adjustment data; Step S32: normalizing the flight state attitude control adjustment data to obtain flight state attitude control normalized data; Step S32: Perform UAV control strategy logic learning based on the flight state attitude control normalization data to obtain the UAV control logic strategy.

8. The drone control method according to claim 7, characterized in that: Step S31 includes the following steps: Step S311: performing simulation calculation of abnormal numerical fluctuations of yaw angle and pitch angle of the UAV flight state data according to the unbalanced decomposition data of rain impact force and the structural fluid load data, and obtaining abnormal numerical fluctuation data of yaw angle and abnormal numerical fluctuation data of pitch angle respectively; Step S312: performing a numerical instantaneous increment continuity analysis on the abnormal yaw angle numerical fluctuation data and the abnormal pitch angle numerical fluctuation data to obtain the abnormal instantaneous increment continuity data of the yaw angle and the abnormal instantaneous increment continuity data of the pitch angle; Step S313: performing linkage coupling effect analysis based on the continuous data of the abnormal instantaneous increment of the yaw angle and the continuous data of the abnormal instantaneous increment of the pitch angle to obtain yaw / pitch coupling effect data; Step S314: performing thrust vector correction on the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data according to the yaw / pitch coupling effect data to obtain flight thrust vector correction data; Step S315: performing central inertia variable compensation correction on the flight thrust vector correction data to obtain central inertia variable compensation data; Step S316: Perform attitude control adjustment on the UAV flight state data according to the flight thrust vector correction data and the central inertia variable compensation data to obtain flight state attitude control adjustment data.

9. The drone control method according to claim 8, characterized in that: Step S314 includes the following steps: According to the yaw / pitch coupling effect data, the yaw angle value abnormal fluctuation data and the pitch angle value abnormal fluctuation data are used to calculate the yaw and pitch flight resistance distribution difference, so as to obtain the yaw / pitch resistance distribution difference data; Calculate the moment balance requirement for the yaw and pitch resistance distribution difference data to obtain the resistance difference moment balance requirement data; Performing thrust vector direction correction processing between yaw angle / pitch angle on yaw and pitch resistance distribution difference data to obtain thrust vector direction correction data; Based on the drag difference torque balance requirement data and the thrust vector direction correction data, the thrust vector distribution is optimized to obtain the thrust vector distribution optimization data; The thrust vector correction is performed according to the drag difference torque balance requirement data, the thrust vector direction correction data and the thrust vector distribution optimization data to obtain the flight thrust vector correction data.

10. A drone control system, characterized in that: Used to execute the drone control method according to claim 1, the drone control system comprises: The data acquisition module is used to collect the real-time data of the UAV flight environment and the flight status through the sensors carried by the UAV, so as to obtain the real-time data of the UAV flight environment and the flight status data of the UAV; extract the real-time wind condition data and rainfall characteristic data from the real-time data of the UAV flight environment, so as to obtain the real-time wind condition data of the flight environment and the rainfall characteristic data of the flight environment respectively; The force imbalance decomposition module is used to evaluate the structural fluid load of the UAV flight status data according to the real-time wind condition data of the flight environment to obtain the structural fluid load data; based on the structural fluid load data, the force imbalance structural decomposition of the flight environment rainfall characteristic data is performed to obtain the rain impact force imbalance decomposition data; The attitude control adjustment module is used to perform attitude control adjustment on the UAV flight state data according to the rain impact force imbalance decomposition data and the structural fluid load data to obtain the flight state attitude control adjustment data; perform UAV control strategy logic learning according to the flight state attitude control adjustment data to obtain the UAV control logic strategy; The execution firmware design module is used to automatically execute the firmware design for the drone control logic strategy, obtain the drone control logic strategy firmware, and embed the drone control logic strategy firmware into the drone control center to execute the drone control method.

Images