Unmanned aerial vehicle attitude control method and system based on controllable turbulence environment

Abstract

The invention relates to the crossing field of unmanned aerial vehicle control and fluid mechanics, in particular to an unmanned aerial vehicle attitude control method and system based on a controllable turbulence environment, and the method comprises the steps: carrying out the attitude response analysis of an unmanned aerial vehicle in different flight modes in the controllable turbulence environment; constructing a turbulence characteristic-attitude response correlation model; a self-adaptive control algorithm based on a turbulence characteristic-attitude response correlation model is introduced into a flight control system, and a self-adaptive control strategy is applied; performing verification and effect evaluation on the adaptive control algorithm through an evaluation experiment; the unmanned aerial vehicle attitude control system is applied to the unmanned aerial vehicle attitude control method, can accurately capture the moment when the unmanned aerial vehicle makes attitude response when encountering the impact of turbulent vortex and track the dynamic change of the attitude, and can effectively reduce the attitude fluctuation amplitude and response time of the unmanned aerial vehicle in complex airflow.

Description

Technical Field

[0001] The present invention relates to the cross - field of unmanned aerial vehicle (UAV) control and fluid mechanics, and particularly to a UAV attitude control method and system based on a controllable turbulent environment. Background Art

[0002] With the increasingly wide application of UAVs in fields such as logistics transportation, disaster monitoring, and security patrol, the turbulent interference encountered by UAVs in the real environment cannot be ignored. The shear vortices and pulsating pressures in a high - turbulence - intensity flow field often cause large fluctuations in the UAV attitude, and even lead to the over - limit or instability of the flight control system. Existing research is usually carried out under relatively stable wind tunnel conditions, lacking attention to the cases of high - turbulence - intensity or multi - scale vortex interference, and it is difficult to reveal the attitude response process of UAVs in real complex airflows.

[0003] Therefore, there is an urgent need for a UAV attitude analysis and stability control method and system that can utilize a controllable turbulent environment to quantitatively measure the relationship between UAV attitude, moment changes, and turbulence intensity under laboratory conditions, and can improve the existing flight control algorithm to adaptively control UAVs in a turbulent environment. Summary of the Invention

[0004] In view of the above problems, the present invention provides a UAV attitude control method and system based on a controllable turbulent environment to solve the technical problems proposed in the above background art.

[0005] The technical solutions adopted by the present invention to solve its technical problems are as follows:

[0006] A UAV attitude control method based on a controllable turbulent environment includes the following steps:

[0007] S100: Construct a controllable turbulent environment, and under the controllable turbulent environment, conduct UAV attitude response analysis for different flight modes;

[0008] S200: Based on the results of the UAV attitude response analysis, construct a "turbulence characteristic - attitude response" correlation model;

[0009] S300: Introduce an adaptive control algorithm based on the "turbulence characteristic - attitude response" correlation model into the flight control system and apply the adaptive control strategy;

[0010] S400: Verify and evaluate the effect of the adaptive control algorithm through an evaluation experiment.

[0011] Further, the UAV attitude response analysis in S100 specifically includes the following steps:

[0012] S110: Set the current flight mode of the UAV;

[0013] S120: Obtain the current training dataset in the current flight mode, where the current training dataset includes IMU data, optical marker pose data, and PIV data under different turbulence environments;

[0014] S130: Extract the attitude fluctuation events of the UAV based on the IMU data and the optical marker pose data, and extract the vortex perturbation events in the turbulent flow field based on the PIV data;

[0015] S140: Synchronize and align the IMU data and the optical marker pose data with the PIV data, and capture the attitude response moments when the UAV's attitude fluctuates due to encountering vortex perturbation events;

[0016] S150: According to the captured attitude response moments, statistically analyze the turbulence characteristics and the UAV attitude response characteristics at different attitude response moments;

[0017] S160: Analyze the influence of the turbulence characteristics on the UAV attitude response characteristics, and determine the instability flow conditions that may cause the output of the flight control system to be saturated or lag and become unstable in the current flight mode;

[0018] S170: Determine whether all current flight modes are set. If so, output the current training dataset and the current instability flow conditions in different current flight modes to obtain the full-mode training dataset and the full-mode instability flow conditions that may cause the flight control system to be unstable; if not, change the current flight mode and repeat S110 - S160.

[0019] Furthermore, the construction of the "turbulence characteristic - attitude response" correlation model in S200 specifically includes the following steps:

[0020] S210: Clean the full-mode training dataset by processing the duplicate data, missing data, and abnormal data in the full-mode training dataset, so as to extract effective turbulence characteristic data and effective UAV attitude response characteristic data;

[0021] S220: Perform parametric preprocessing on the effective turbulence characteristic data and the effective UAV attitude response characteristic data to obtain effective turbulence characteristic parameters and effective UAV attitude response characteristic parameters;

[0022] S230: Construct a "turbulence characteristic - attitude response" correlation model with the turbulence characteristic parameters as independent variables and the UAV attitude response characteristic parameters as dependent variables, based on a multiple regression model or a non-linear model based on physical mechanisms;

[0023] S240: Input the turbulence characteristic parameters into the "turbulence characteristic - attitude response" correlation model, and the "turbulence characteristic - attitude response" correlation model outputs the corresponding estimated UAV attitude response characteristic parameters;

[0024] S250: Compare the estimated parameters of the UAV attitude response characteristics with the effective UAV attitude response characteristic parameters to verify the accuracy of the "turbulence characteristic - attitude response" correlation model;

[0025] S260: Output the "turbulence characteristic - attitude response" correlation model.

[0026] Furthermore, the adaptive control algorithm introduced in S300 specifically includes the following steps:

[0027] S310: Input the full - mode instability flow conditions that may cause the flight control system to be unstable into the flight control system;

[0028] S320: Input the IMU data and turbulence estimation collected during the UAV flight into the flight control system, where the turbulence estimation includes turbulence intensity estimation, eddy scale estimation, and eddy frequency estimation;

[0029] S330: Classify the collected turbulence estimations into different turbulence levels according to the magnitudes of the turbulence intensity estimation, eddy scale estimation, eddy frequency estimation, and whether the full - mode instability flow conditions are reached, and execute S340 and S360 respectively;

[0030] S340: The feed - forward controller in the flight control system predicts different disturbance signals for the turbulence estimation data of different turbulence levels;

[0031] S350: The feed - forward compensation controller generates a feed - forward compensation signal;

[0032] S360: Based on the "turbulence characteristic - attitude response" correlation model, the gain - scheduling controller in the flight control system calculates different initial control gains for the turbulence estimation data of different turbulence levels;

[0033] S370: The gain - scheduling controller receives the feed - forward compensation signal and calculates the compensation control gain. The initial control gain is superimposed with the compensation control gain to obtain the final control gain, and adaptive gain scheduling is performed on the UAV;

[0034] S380: The flight control system outputs flight control instructions.

[0035] Furthermore, S400 specifically includes the following steps:

[0036] S401: Conduct test preparations and check the UAV flight control system and various sensors used to collect inertial field data and flow field data;

[0037] S402: Divide the experimental environment into N different turbulence conditions, set the current experimental environment as the nth turbulence condition, where n = 1;

[0038] S403: Set the control algorithm in the flight control system to the original algorithm;

[0039] S404: The UAV conducts original flight tests under different flight modes respectively, and completes multiple original flight experiments;

[0040] S405: Record and output the original data of the UAV attitude response characteristics in the current experimental environment obtained from multiple original flight experiments;

[0041] S406: Set the control algorithm in the flight control system to the current adaptive control algorithm;

[0042] S407: The UAV conducts optimized flight tests under different flight modes respectively, and completes multiple optimized flight experiments;

[0043] S408: Record and output the optimized data of the UAV attitude response characteristics in the current experimental environment obtained from multiple optimized flight experiments;

[0044] S409: Determine whether n is greater than or equal to N; if so, go to S410; if not, let n = n + 1, set the current experimental environment to the nth kind of turbulence condition, and repeat S403 - S408;

[0045] S410: Analyze the optimization effect of the adaptive control algorithm relative to the original algorithm by comparing the initial data of the UAV attitude response characteristics with the optimized data of the UAV attitude response characteristics;

[0046] S411: Determine whether the optimization effect reaches the expected effect; if so, determine the current adaptive control algorithm as the final adaptive control algorithm, and introduce the final adaptive control algorithm into the flight control system; if not, further optimize the adaptive control algorithm, and repeat S401 - S410.

[0047] A UAV attitude control system based on a controllable turbulence environment, which is applied to the above-mentioned UAV attitude control method based on a controllable turbulence environment, includes:

[0048] Controllable turbulence environment construction module: The controllable turbulence environment construction module is used to provide a complex flow field with multi-scale turbulence that is close to the real environment for the UAV indoors;

[0049] UAV attitude analysis module: The UAV attitude analysis module is used to analyze the instantaneous attitude response of the UAV when it encounters the impact of a turbulence vortex;

[0050] Correlation model construction module: The correlation model construction module is used to construct a "turbulence feature - attitude response" correlation model with the turbulence feature parameters as independent variables and the UAV attitude response feature parameters as dependent variables by using the full-mode training data set;

[0051] Adaptive control module: The adaptive control module is used to control the attitude of the unmanned aerial vehicle (UAV) in a turbulent environment.

[0052] Experimental verification module: The experimental verification module is used to verify and evaluate the adaptive control algorithm under different turbulent environments and different flight modes.

[0053] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0054] 1. A UAV attitude control method based on a controllable turbulent environment provided by the present invention can comprehensively analyze the response of the UAV attitude to high-turbulence and multi-scale vortex structures in an indoor controllable turbulent environment. Through the synchronization and alignment of UAV IMU data and PIV flow field data, the moment when the UAV makes an attitude response when encountering the impact of a turbulent vortex can be accurately captured, and the dynamic change of the attitude can be tracked.

[0055] 2. A UAV attitude control method based on a controllable turbulent environment provided by the present invention can effectively reduce the attitude fluctuation amplitude and response time of the UAV in complex airflows by constructing a "turbulence characteristic - attitude response" correlation model and introducing an adaptive control algorithm into the flight control system.

[0056] 3. A UAV attitude control system based on a controllable turbulent environment provided by the present invention, which is applied to a UAV attitude control method based on a controllable turbulent environment, can provide reference and technical reserves for subsequent UAV anti-disturbance research and industrial applications in larger scales and complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0058] Figure 1 It is a schematic flow chart of the UAV attitude control method based on a controllable turbulent environment described in the present invention;

[0059] Figure 2 For Figure 1 the flow schematic diagram of S100 in;

[0060] Figure 3 For Figure 1 the flow schematic diagram of S200 in;

[0061] Figure 4 For Figure 1 the flow schematic diagram of S300 in;

[0062] Figure 5 For Figure 1 the flow schematic diagram of S400 in

[0063] Figure 6 the structural schematic diagram of the UAV attitude control system based on the controllable turbulence environment described in the present invention. Specific implementation manners

[0064] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0065] The present invention provides a UAV attitude control method based on a controllable turbulence environment, as Figures 1-5 shown. The UAV attitude control method based on a controllable turbulence environment includes the following steps:

[0066] S100: Construct a controllable turbulence environment, and under the controllable turbulence environment, perform UAV attitude response analysis for different flight modes;

[0067] Among them, the constructed controllable turbulence environment can be regulated within different turbulence ranges. By constructing the controllable turbulence environment, the UAV can be subjected to attitude response analysis under different turbulence environments, and the attitude response of the UAV in the face of a complex turbulence environment can be obtained.

[0068] Specifically, the UAV attitude response analysis process described in S100 of the embodiment of the present invention includes the following steps:

[0069] S110: Set the current flight mode of the UAV;

[0070] Among them, the optional current flight modes include a hover mode, a climb mode, and a level flight mode.

[0071] S120: Obtain the current training data set in the current flight mode. The current training data set includes IMU data, optical marker pose data, and PIV data under different turbulence environments;

[0072] Among them, the IMU data is collected by mounting a flight control system with an open interface on the UAV. The IMU data is collected in real time by sensors such as accelerometers, gyroscopes, and magnetic compasses, and the sampling frequency range is 100 - 500 Hz; the optical marker pose data is obtained by a high-precision optical marker and a high-speed camera configured outside the test area, and the acquisition frame rate is 50 - 100 Hz; the PIV data is measured in the test area by particle image velocimetry (abbreviation: PIV) and a hot-wire anemometer. The PIV data includes flow field velocity, turbulence intensity distribution, vortex core diameter, and turbulence energy spectrum information.

[0073] S130: Extract the attitude fluctuation events of the UAV according to the IMU data and the optical marker pose data, and extract the vortex perturbation events in the turbulent flow field according to the PIV data;

[0074] Among them, the attitude fluctuation event refers to an event in which the body attitude of the UAV or the torque of the UAV rotor suddenly changes, and the vortex perturbation event refers to an event in which a high-speed vortex in the turbulent flow field is disturbed after hitting an external object.

[0075] S140: Synchronize and align the IMU data, the optical marker pose data, and the PIV data, and capture the attitude response moment when the UAV has an attitude fluctuation due to encountering a vortex perturbation event;

[0076] Among them, the IMU data, the optical marker pose data, and the PIV data are synchronized in time and aligned in spatial distribution. By synchronizing and aligning the UAV IMU data, the optical marker pose data, and the PIV data, the attitude response moment of the UAV can be accurately captured and the dynamic changes of the UAV attitude can be tracked.

[0077] S150: According to the captured attitude response moments, statistically analyze the turbulence characteristics and the UAV attitude response characteristics at different attitude response moments;

[0078] Among them, the turbulence characteristics include turbulence intensity, vortex scale, and vortex frequency, and the UAV attitude response characteristics include the UAV attitude deviation amplitude, the UAV rotor torque pulsation, and the UAV attitude recovery time.

[0079] S160: Analyze the influence of the turbulence characteristics on the UAV attitude response characteristics, and determine the unstable flow conditions that may cause the output of the flight control system to be saturated or lag and become unstable in the current flight mode;

[0080] Among them, the current training data set includes turbulence feature data and the corresponding UAV attitude response feature data; due to the limitations of the UAV hardware conditions, there is a design upper limit for the output of the flight control system. When the output of the flight control system reaches the design upper limit and saturation occurs, the saturation phenomenon will damage the performance of the flight control system, making it difficult for the UAV to perform precise control as expected, resulting in the UAV tending to be unstable; when the output of the flight control system cannot respond immediately to the input signal but has a certain time delay, a lag phenomenon will occur. The lag phenomenon will cause the response speed of the flight control system to slow down, reducing the maneuverability and stability of the UAV.

[0081] S170: Determine whether all the current flight modes are set up. If so, output the current training data set and the current instability flow conditions under different current flight modes to obtain the full-mode training data set and the full-mode instability flow conditions that may cause the flight control system to be unstable; if not, replace the current flight mode and repeat S110 - S160.

[0082] S200: Based on the UAV attitude response analysis results, construct a "turbulence feature - attitude response" correlation model;

[0083] Among them, the "turbulence feature - attitude response" correlation model is the correspondence between the turbulence features under different turbulence environments and the attitude response features of the UAV under the corresponding turbulence environments.

[0084] Specifically, the construction of the "turbulence feature - attitude response" correlation model in S200 of the embodiment of the present invention includes the following steps:

[0085] S210: Clean the full-mode training data set by processing the duplicate data, missing data, and abnormal data in the full-mode training data set, so as to extract effective turbulence feature data and effective UAV attitude response feature data;

[0086] Among them, the process of cleaning the full-mode training data set includes: deleting duplicate data, filling in missing data by interpolation method, and deleting or correcting abnormal data.

[0087] S220: Perform parametric preprocessing on the effective turbulence feature data and the effective UAV attitude response feature data to obtain effective turbulence feature parameters and effective UAV attitude response feature parameters;

[0088] Among them, through parametric preprocessing, the effective turbulence feature data and the effective UAV attitude response feature data are scaled to a unified range to solve the problem of dimensional difference between the effective turbulence feature data and the effective UAV attitude response feature data.

[0089] S230: Taking the turbulence characteristic parameters as independent variables and the UAV attitude response characteristic parameters as dependent variables, construct a "turbulence characteristic - attitude response" correlation model based on a multiple regression model or a non - linear model based on physical mechanisms;

[0090] S240: Input the

[0091] turbulence characteristic parameters into the "turbulence characteristic - attitude response" correlation model, and the "turbulence characteristic - attitude response" correlation model outputs the corresponding UAV attitude response characteristic prediction parameters;

[0092] Among them, the UAV attitude response characteristic prediction parameters are predicted by the "turbulence characteristic - attitude response" correlation model, and the UAV attitude response characteristic prediction parameters can predict the upcoming attitude response of the UAV in a complex turbulence environment for the flight control system.

[0093] S250: Compare the UAV attitude response characteristic prediction parameters with the effective UAV attitude response characteristic parameters to verify the accuracy of the "turbulence characteristic - attitude response" correlation model;

[0094] Among them, the effective turbulence characteristic parameters corresponding to the effective UAV attitude response characteristic parameters extracted from the full - mode training dataset are consistent with the turbulence characteristic parameters input in the correlation model. If the mean square error between the UAV attitude response characteristic prediction parameters and the effective UAV attitude response characteristic parameters is within the expected range, the accuracy of the "turbulence characteristic - attitude response" correlation model meets the requirements.

[0095] S260: Output the "turbulence characteristic - attitude response" correlation model.

[0096] Among them, the output "turbulence characteristic - attitude response" correlation model is presented in the form of the established "turbulence characteristic - attitude response" function or chart.

[0097] S300: Introduce an adaptive control algorithm based on the "turbulence characteristic - attitude response" correlation model into the flight control system and apply the adaptive control strategy;

[0098] Among them, due to the fact that in a turbulence environment, the dynamic characteristics of the UAV flight control system change at all times. Therefore, it is necessary to establish a more accurate flight control system control strategy. The present invention introduces an improved adaptive control algorithm into the flight control system, thereby applying the adaptive control strategy in the flight control system to effectively reduce the attitude fluctuation amplitude and response time of the UAV in complex airflows; the flight control system includes a gain - scheduling controller and a feed - forward controller. The adaptive control algorithm combines the adaptive gain - scheduling method and the feed - forward compensation method, enabling the flight control system to perform accurate attitude control under different turbulence conditions and ensuring the stability of the UAV under different turbulence conditions.

[0099] Specifically, the adaptive control algorithm described in this embodiment includes the following steps:

[0100] S310: Input the full-mode instability flow conditions that may cause the flight control system to be unstable into the flight control system;

[0101] S320: Input the IMU data and turbulence estimation collected during the flight of the UAV into the flight control system. The turbulence estimation includes turbulence intensity estimation, eddy scale estimation, and eddy frequency estimation;

[0102] Among them, the basic flow field data of the external flow field of the UAV is obtained through a hot-wire anemometer carried on the UAV, and the external flow field is numerically simulated through the built-in flow field simulation software of the UAV, so as to obtain the turbulence estimation.

[0103] S330: Classify the collected turbulence estimations into different turbulence levels according to the magnitudes of the turbulence intensity estimation, eddy scale estimation, and eddy frequency estimation and whether the full-mode instability flow conditions are reached, and execute S340 and S360 respectively;

[0104] Among them, the higher the magnitudes of the turbulence intensity estimation, eddy scale estimation, and eddy frequency estimation, the higher the turbulence level of the corresponding turbulence estimation. The turbulence estimation that reaches the full-mode instability flow conditions is the highest turbulence level; classifying the turbulence estimations into different turbulence levels and applying different control gains and feedforward compensations to the turbulence estimations of different turbulence levels enables the UAV to obtain precise attitude control under different turbulence conditions.

[0105] S340: The feedforward controller in the flight control system predicts different disturbance signals for the turbulence estimation data of different turbulence levels;

[0106] Among them, the UAV flight control system is disturbed in a complex turbulence environment. The disturbances include changes in the external flow field and changes in the dynamic load of the UAV. The feedforward controller can predict these upcoming disturbances based on the turbulence estimation data.

[0107] S350: The feedforward compensation controller generates a feedforward compensation signal;

[0108] Among them, the feedforward compensator calculates corresponding control measures based on the predicted upcoming disturbances and generates the control measures into a feedforward compensation signal.

[0109] S360: Based on the "turbulence characteristics - attitude response" correlation model, the gain scheduling controller in the flight control system calculates different initial control gains for the turbulence estimation data of different turbulence levels;

[0110] Among them, the gain-scheduling controller includes multiple local controllers. The gain-scheduling controller designs a gain-scheduling algorithm based on the "turbulence characteristics - attitude response" correlation model. The gain-scheduling controller receives real-time turbulence estimation data of different turbulence levels, calculates the parameters of each local controller according to the gain-scheduling algorithm, generates local control gains, and thus obtains the global control gain of the unmanned aerial vehicle (UAV). This global control gain is the initial control gain. By flexibly scheduling the control gains of the local controllers, the gain-scheduling controller can achieve gain scheduling under different turbulence conditions of the flight control system, ensuring the flexibility of the flight control system.

[0111] S370: The gain-scheduling controller receives the feedforward compensation signal and calculates the compensation control gain. The initial control gain is superimposed with the compensation control gain to obtain the final control gain, and adaptive gain scheduling is performed on the UAV.

[0112] Among them, since the robustness of the gain-scheduling algorithm is affected when facing unknown disturbances, by superimposing the feedforward compensation gain and the initial control gain, the robustness of the gain-scheduling algorithm can be significantly enhanced, ensuring that the flight control system can flexibly, accurately, and stably perform attitude control on the UAV in a complex turbulence environment.

[0113] S380: The flight control system outputs flight control instructions.

[0114] S400: Verify and evaluate the adaptive control algorithm through an evaluation experiment.

[0115] Among them, to verify the generality and robustness of the adaptive control algorithm, the present invention designs the evaluation experiment. The evaluation experiment continuously optimizes the adaptive control algorithm by evaluating the optimization effect of the adaptive control algorithm on the UAV under different turbulence environments and different flight modes.

[0116] Specifically, step S400 of the embodiment of the present invention includes the following steps:

[0117] S401: Conduct test preparations, and check the UAV flight control system and various sensors for collecting inertial field data and flow field data.

[0118] Among them, the sensors for collecting the inertial field data are various sensors for collecting IMU data, and the sensors for collecting the flow field data are various sensors for collecting turbulence estimation data.

[0119] S402: Divide the experimental environment into N different turbulence conditions, set the current experimental environment as the nth turbulence condition, and n = 1.

[0120] Among them, different turbulence conditions include the external flow field conditions where the UAV is located and the UAV load conditions. Conducting tests under N different turbulence conditions helps to comprehensively evaluate the generality and robustness of the adaptive control algorithm.

[0121] S403: Set the control algorithm in the flight control system as the original algorithm;

[0122] Among them, the original algorithm is the control algorithm before the adaptive control algorithm is introduced into the UAV flight control system. In the process of building the model of the original algorithm, the attitude response analysis of the UAV in a complex turbulence environment is lacking. Taking the control effect of the original algorithm as the benchmark, by comparing the optimization level of the control effect of the adaptive control algorithm relative to the original algorithm, the adaptive control algorithm is verified.

[0123] S404: The UAV conducts original flight tests under different flight modes respectively to complete multiple original flight experiments;

[0124] S405: Record and output the original data of the UAV attitude response characteristics in the current experimental environment obtained from multiple original flight experiments;

[0125] S406: Set the control algorithm in the flight control system as the current adaptive control algorithm;

[0126] S407: The UAV conducts optimized flight tests under different flight modes respectively to complete multiple optimized flight experiments;

[0127] Among them, the different flight modes tested in the optimized flight experiment correspond one by one to the different flight modes tested in the original flight experiment.

[0128] S408: Record and output the optimized data of the UAV attitude response characteristics corresponding to different flight modes in the current experimental environment obtained from multiple optimized flight experiments;

[0129] S409: Determine whether n is greater than or equal to N; if so, go to S410; if not, let n = n + 1, set the current experimental environment as the nth turbulence condition, and repeat S403 - S408;

[0130] S410: By comparing the original data of the UAV attitude response characteristics with the optimized data of the UAV attitude response characteristics, analyze the optimization effect of the adaptive control algorithm relative to the original algorithm;

[0131] S411: Determine whether the optimization effect reaches the expected effect; if so, determine the current adaptive control algorithm as the final adaptive control algorithm and introduce the final adaptive control algorithm into the flight control system; if not, further optimize the adaptive control algorithm and repeat S401 - S410.

[0132] Among them, if the amplitude of the UAV attitude deviation is reduced by 20%-30% on average compared to the amplitude of the UAV attitude deviation, it indicates that the adaptive control algorithm has a more significant stabilization effect than the original algorithm in a complex turbulent environment, that is, the optimization effect meets the expectations.

[0133] The present invention also provides a UAV attitude control system based on a controllable turbulent environment, which applies the above-mentioned UAV attitude control method based on a controllable turbulent environment. As Figure 6 shown, the UAV attitude control system based on a controllable turbulent environment includes:

[0134] Controllable turbulent environment construction module: The controllable turbulent environment construction module is used to provide a complex flow field with multi-scale turbulence close to the real environment for the UAV indoors, making the attitude analysis and stability control of the UAV more realistic;

[0135] UAV attitude analysis module: The UAV attitude analysis module is used to analyze the instantaneous attitude response of the UAV when it encounters the impact of turbulent vortices, so as to identify the instability mode of the UAV in the hover, climb or level flight mode, and examine the influence of turbulence characteristics such as turbulence intensity and vortex scale on the amplitude of the UAV attitude deviation and the recovery time;

[0136] Correlation model construction module: The correlation model construction module is used to use the full-mode training data set to construct a "turbulence characteristic-attitude response" correlation model with the turbulence characteristic parameters as independent variables and the UAV attitude response characteristic parameters as dependent variables, providing a basic model for the construction of the adaptive control algorithm;

[0137] Adaptive control module: The adaptive control module is used to control the attitude of the UAV in a turbulent environment. The adaptive control module introduces the adaptive scheduling control algorithm and the feedforward compensation algorithm into the flight control system, which can effectively reduce the amplitude of the UAV attitude fluctuation and the response time in complex airflows, and can perform more stable adaptive control on the UAV in a complex turbulent environment;

[0138] Experimental verification module: The experimental verification module is used to verify and evaluate the adaptive control algorithm in different turbulent environments and different flight modes to ensure the generality and robustness of the adaptive control algorithm.

[0139] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for attitude control of an unmanned aerial vehicle based on a controllable turbulent environment, characterized in that: The steps include: S100: Construct a controllable turbulence environment and perform attitude response analysis of UAVs in different flight modes in the controllable turbulence environment; S200: Based on the analysis results of the UAV attitude response, a "turbulence characteristics-attitude response" correlation model is constructed; S300: Introduce an adaptive control algorithm based on the "turbulence characteristics-attitude response" correlation model in the flight control system and apply an adaptive control strategy; S400: Verify and evaluate the effectiveness of the adaptive control algorithm through evaluation experiments.

2. The method for controlling the attitude of an unmanned aerial vehicle based on a controllable turbulent environment according to claim 1, characterized in that: The drone attitude response analysis in S100 specifically includes the following steps: S110: Set the current flight mode of the drone; S120: Acquire a current training data set in the current flight mode, where the current training data set includes IMU data, optical marker position data, and PIV data in different turbulence environments; S130: extracting the attitude fluctuation events of the UAV according to the IMU data and the optical marker posture data, and extracting the vortex disturbance events in the turbulent flow field according to the PIV data; S140: Synchronize and align the IMU data and the optical marker pose data with the PIV data to capture the attitude response moment when the attitude of the UAV fluctuates due to the vortex disturbance event; S150: according to the captured attitude response moments, statistics are collected on turbulence characteristics and attitude response characteristics of the UAV at different attitude response moments; S160: Analyze the impact of turbulence characteristics on the attitude response characteristics of the UAV, and determine the unstable flow conditions that may cause the flight control system output to be saturated or hysteresis in the current flight mode and become unstable; S170: Determine whether all settings of the current flight mode have been completed. If so, output the current training data set and the current unstable flow conditions under different current flight modes to obtain the full-mode training data set and the full-mode unstable flow conditions that may cause instability of the flight control system; if not, change the current flight mode and repeat S110-S160.

3. The method for controlling the attitude of an unmanned aerial vehicle based on a controllable turbulent environment according to claim 2, characterized in that: The construction of the "turbulence characteristics-attitude response" correlation model in S200 specifically includes the following steps: S210: cleaning the full-mode training data set by processing duplicate data, missing data and abnormal data in the full-mode training data set, thereby extracting effective turbulence feature data and effective UAV attitude response feature data; S220: performing parameter preprocessing on the effective turbulence characteristic data and the effective UAV attitude response characteristic data, so as to obtain effective turbulence characteristic parameters and effective UAV attitude response characteristic parameters; S230: Taking turbulence characteristic parameters as independent variables and UAV attitude response characteristic parameters as dependent variables, a "turbulence characteristic-attitude response" correlation model is constructed based on a multivariate regression model or a nonlinear model based on physical mechanisms; S240: inputting turbulence characteristic parameters into the "turbulence characteristic-attitude response" association model, and the "turbulence characteristic-attitude response" association model outputs corresponding effective UAV attitude response characteristic estimation parameters; S250: Compare the estimated parameters of the UAV attitude response characteristics with the effective UAV attitude response characteristic parameters to verify the accuracy of the "turbulence characteristics-attitude response" correlation model; S260: Output the "turbulence characteristics-attitude response" correlation model.

4. The method for controlling the attitude of an unmanned aerial vehicle based on a controllable turbulent environment according to claim 3, characterized in that: The adaptive control algorithm introduced in S300 specifically includes the following steps: S310: Inputting full-mode unstable flow conditions that may cause instability of the flight control system into the flight control system; S320: inputting the IMU data and turbulence estimation collected during the flight of the UAV into the flight control system, wherein the turbulence estimation includes turbulence intensity estimation, eddy scale estimation and eddy frequency estimation; S330: classifying the collected turbulence estimates into different turbulence levels according to the estimated turbulence intensity, the estimated eddy scale, the estimated eddy frequency, and whether the full-mode instability flow condition is reached, and executing S340 and S360 respectively; S340: a feedforward controller in the flight control system predicts different disturbance signals according to turbulence estimation data of different turbulence levels; S350: A feedforward compensation controller generates a feedforward compensation signal; S360: Based on the "turbulence characteristics-attitude response" association model, the gain scheduling controller in the flight control system calculates different initial control gains for turbulence estimation data of different turbulence levels; S370: The gain scheduling controller receives the feedforward compensation signal and calculates the compensation control gain, the initial control gain is superimposed on the compensation control gain to obtain the final control gain, and adaptive gain scheduling is performed on the UAV; S380: The flight control system outputs flight control instructions.

5. The method for controlling the attitude of an unmanned aerial vehicle based on a controllable turbulent environment according to claim 4, characterized in that: The S400 specifically includes the following steps: S401: Prepare for the test and check the UAV flight control system and various sensors used to collect inertial field data and flow field data; S402: Divide the experimental environment into N different turbulence conditions, and set the current experimental environment to the nth turbulence condition, where n=1; S403: Setting the control algorithm in the flight control system to the original algorithm; S404: The UAV conducts original flight tests in different flight modes and completes multiple original flight experiments; S405: Record and output the original data of the attitude response characteristics of the UAV in the current experimental environment obtained from multiple original flight experiments; S406: Setting the control algorithm in the flight control system to the current adaptive control algorithm; S407: The UAV performs optimized flight tests in different flight modes and completes multiple optimized flight experiments; S408: Record and output the optimization data of the attitude response characteristics of the UAV in the current experimental environment obtained by multiple optimization flight experiments; S409: Determine whether n is greater than or equal to N; if so, go to S410; If not, set n=n+1, set the current experimental environment to the nth turbulence condition, and repeat S403-S408; S410: Analyze the optimization effect of the adaptive control algorithm relative to the original algorithm by comparing the initial data of the attitude response characteristics of the UAV with the optimized data of the attitude response characteristics of the UAV; S411: Determine whether the optimization effect reaches the expected effect; if so, determine that the current adaptive control algorithm is the final adaptive control algorithm, and introduce the final adaptive control algorithm into the flight control system; if not, further optimize the adaptive control algorithm, and repeat S401-S410.

6. A UAV attitude control system based on a controllable turbulent environment, applied to a UAV attitude control method based on a controllable turbulent environment as described in any one of claims 1 to 5, characterized in that: include: Controllable turbulence environment building module: The controllable turbulence environment building module is used to provide a complex flow field with multi-scale turbulence close to the real environment for the UAV indoors; UAV attitude analysis module: The UAV attitude analysis module is used to analyze the instantaneous attitude response of the UAV when it encounters the impact of the turbulent vortex; Correlation model building module: The correlation model building module is used to build a "turbulence characteristic-attitude response" correlation model using a full-mode training data set, taking turbulence characteristic parameters as independent variables, and taking UAV attitude response characteristic parameters as dependent variables; Adaptive control module: The adaptive control module is used to perform attitude control on the UAV in a turbulent environment; Experimental verification module: The experimental verification module is used to verify and evaluate the effect of the adaptive control algorithm in different turbulence environments and different flight modes.

Images