Outdoor test flight verification platform for intelligent fault-tolerant flight control system of fixed-wing UAV

Abstract

The present invention is an outdoor test flight verification platform for an intelligent fault-tolerant flight control system for fixed-wing unmanned aerial vehicles, which relates to the field of unmanned aerial vehicles. The platform is divided into an unmanned aerial vehicle platform and a ground platform. The unmanned aerial vehicle platform includes: a fixed-wing unmanned aerial vehicle aircraft model, which is used to carry equipment required for flight; a flight control board, which is used to measure the state information of the unmanned aerial vehicle and control the flight; an airborne industrial control computer, which is used to carry the control algorithm, calculate and output the fault-tolerant control law to the flight control board; an airborne data transmission radio, which is used for communication between the unmanned aerial vehicle and the ground. The ground platform includes: a ground data transmission radio, which is used for communication between the unmanned aerial vehicle and the ground; a Mission Planner ground station, which is used to display the state information of the unmanned aerial vehicle and flight parameters; and a fault-tolerant ground station, which is used to display the output results of the control algorithm and the algorithm parameters. The unmanned aerial vehicle outdoor test flight verification platform of the present invention can intuitively display the flight state of the unmanned aerial vehicle in an actual environment when a failure occurs, accumulate test flight experience, and reduce the cost of test flight trial and error.

Description

Technical Field

[0001] The present application relates to intelligent fault-tolerant flight control technology for fixed-wing UAVs, and in particular to an outdoor flight test verification platform. Background Art

[0002] Fixed-wing UAVs play an important role in surveying and mapping, geology, petroleum, agriculture and forestry due to their advantages such as long flight time, high flight speed and strong carrying capacity. However, when performing complex tasks, fixed-wing UAVs may encounter unpredictable problems such as actuator failures, which limit their reliability and efficiency in changing environments. In order to improve the fault tolerance of UAVs, fault-tolerant control technology has become an important research direction. Fault-tolerant control design can ensure that UAVs can still maintain a certain flight performance when encountering faults, reducing the risk of inefficiency or mission failure caused by faults.

[0003] In previous research on drones, simulation software such as Matlab and Simulink were mainly used to perform numerical simulation on the models. However, there is often a large gap between numerical simulation and the real environment. Test flights in a real environment can more accurately demonstrate the physical characteristics of drones, including aerodynamics, structural strength, and material properties. It can provide actual physical feedback, including environmental factors such as wind speed and air density changes, which are difficult to accurately simulate in digital simulations. By test flying in a real environment, some problems that are ignored in numerical simulations can be discovered. Summary of the invention

[0004] In response to the above-mentioned problems and technical needs, the applicant has proposed an outdoor test flight verification platform for an intelligent fault-tolerant flight control system for fixed-wing UAVs. Taking into account economy, safety and practicality, an outdoor test flight verification platform for an intelligent fault-tolerant flight control system for fixed-wing UAVs has been built. The platform can demonstrate the flight performance of fixed-wing UAVs in real environments and verify the intelligent fault-tolerant control algorithm.

[0005] The main technical solutions adopted in this application are as follows:

[0006] An outdoor test flight verification platform for a fixed-wing UAV intelligent fault-tolerant flight control system, characterized in that the platform comprises: a fixed-wing UAV aircraft model, a digital radio, a flight control board, an airborne industrial computer, a Mission Planner ground station and a fault-tolerant ground station; wherein,

[0007] The fixed-wing UAV aircraft model is used to carry a flight control board, an airborne industrial computer, an airborne data transmission radio and sensor equipment;

[0008] The data transmission radio is divided into an airborne data transmission radio and a ground data transmission radio. The airborne data transmission radio is used to obtain the drone status information in the flight control board, send debugging parameters to the flight control board, and communicate with the ground data transmission radio. The ground data transmission radio is used to provide the Mission Planner ground station with drone status information, obtain the parameters to be uploaded by the Mission Planner ground station and the fault-tolerant ground station, and communicate with the airborne data transmission radio. The data transmission radios communicate with each other through the Serial port.

[0009] The flight control board is used to measure the state information of the UAV during flight and control the flight of the UAV, convert the control law into the PWM value of the steering gear, output the PWM value to the steering gear, and control the flight of the UAV; the flight control board exchanges information with the airborne data transmission radio and the airborne industrial computer;

[0010] The onboard industrial computer is used to carry an intelligent fault-tolerant control algorithm, obtain the drone status information in the flight control board, calculate the fault-tolerant control law, output the fault-tolerant control law, and send it to the flight control board;

[0011] The Mission Planner ground station is used to display the UAV flight information and UAV flight parameters, and obtain the UAV status information through the data transmission radio;

[0012] The fault-tolerant ground station is used to display the fault-tolerant control law, control parameters and intelligent fault-tolerant control algorithm adjustment parameters output by the intelligent fault-tolerant control algorithm, and cooperate with the Mission Planner ground station to adjust the intelligent fault-tolerant control algorithm parameters.

[0013] Furthermore, the drone status information includes attitude angle, attitude angular velocity and flight altitude.

[0014] Furthermore, the digital radio is used for outdoor flight tests and data transmission, and the serial port part uses the Serial serial port library in Python for data interaction; wherein,

[0015] The airborne data transmission radio is used for two-way data transmission with the flight control board, obtaining the UAV status information in the flight control board, and sending flight debugging parameters to the flight control board; the airborne data transmission radio is also used for communication with the ground data transmission radio, transmitting the UAV status information to the Mission Planner ground station for adjustment according to the designed parameters; the airborne data transmission radio is also used for two-way interaction with the airborne industrial computer, obtaining the intelligent fault-tolerant control data in the airborne industrial computer, and sending the intelligent fault-tolerant control debugging parameters to the airborne industrial computer.

[0016] Furthermore, the flight control board has built-in high-precision inertial measurement unit IMU, barometer, GPS module and auxiliary sensors, which are used to collect and process the flight data of the drone in real time; the flight control board is equipped with a flight control program for the normal flight of the drone under fault-free conditions. When the pilot manually injects a fault, the flight control board receives the fault information, causing the drone to generate a corresponding fault, and immediately switches the normal flight control program to the fault-tolerant flight control program in the onboard industrial computer.

[0017] Furthermore, the Mission Planner ground station is used for initial parameter setting, servo neutral position calibration and flight route setting before the UAV takes off. The UAV is shaken in stabilization mode before takeoff, and waypoints are marked on the MissionPlanner map to achieve level flight, ascent, descent and turning of the UAV. The Mission Planner ground station is used for displaying flight data during the flight of the UAV.

[0018] Furthermore, the fault-tolerant ground station is developed based on Python's Pyqt library, and displays the digital display and curve display of fault-tolerant control laws and intelligent fault-tolerant control related data in real time, and uploads intelligent fault-tolerant control related parameters.

[0019] Furthermore, the intelligent fault-tolerant control algorithm is specifically as follows:

[0020] S1 establishes an affine mathematical model of a single-channel attitude loop of a UAV under fault consideration, wherein the attitude loop affine mathematical model includes a UAV fault model;

[0021] Based on the UAV fault model, S2 designs a radial basis function neural network (RBFNN) to fit and estimate fault-related items and design the control law.

[0022] Furthermore, the attitude ring affine mathematical model is:

[0023] (1);

[0024] in, and are attitude angle and attitude angular velocity respectively, and is the aerodynamic parameter, is the control efficiency factor, is the control signal deviation caused by the fault, is an unknown disturbance; is the control law;

[0025] Will as a fault related item.

[0026] Furthermore, the S2 is specifically:

[0027] Set the sliding mode:

[0028] (2);

[0029] in, is the attitude angle tracking deviation, is the fractional order, D is a fractional calculus operator, is a symbolic function; is the desired attitude angle;

[0030] By taking the derivative of (2), we can obtain:

[0031] (3);

[0032] Choose an exponential reaching law:

[0033] (4);

[0034] Among them, is the parameter to be designed;

[0035] The control law is designed as follows:

[0036] (5);

[0037] in, for An estimated value of

[0038] Design RBF neural network for fault related items Perform fitting and design the following RBF neural network:

[0039] (6);

[0040] in, is the estimated value of the ideal network weight, is the Gaussian kernel function of the network, which is as follows:

[0041] (7);

[0042] in, is the input of the network, and are the parameters to be designed;

[0043] The control law Passed to Butterworth low-pass filter, we get ,Right now

[0044] (8);

[0045] Among them, is the filtering error.

[0046] The beneficial technical effects of this application are:

[0047] The present application discloses an outdoor test flight verification platform for an intelligent fault-tolerant flight control system for fixed-wing UAVs. The platform can replace different types of UAVs and fault-tolerant control algorithms according to needs, and the platform is flexible. Through the platform, the flight posture of the UAV in the air and the effect of the fault-tolerant control algorithm can be intuitively understood. By analyzing the flight data stored in each flight, researchers can find some problems in the flight and optimize the UAV parameters and fault-tolerant control algorithm. The platform accumulates experience in outdoor test flights, shortens the development cycle of outdoor test flight fault-tolerant control systems, and reduces the cost of test flights and trials. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a structural block diagram of an outdoor test flight verification platform for an intelligent fault-tolerant flight control system for fixed-wing UAVs;

[0049] Figure 2 It is a design flow chart of an outdoor test flight verification platform for an intelligent fault-tolerant flight control system for fixed-wing UAVs;

[0050] Figure 3 It is the terminal flow chart of the operation of intelligent fault-tolerant control algorithm;

[0051] Figure 4 It is the flight control terminal flow chart;

[0052] Figure 5 This is a schematic diagram of an X-shaped tail rudder;

[0053] Figure 6 This is a schematic diagram of the route setting. DETAILED DESCRIPTION

[0054] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.

[0055] Embodiment 1:

[0056] This application discloses an outdoor test flight verification platform for intelligent fault-tolerant flight control systems for fixed-wing UAVs. Please refer to Figure 1 ,In general, the UAV platform includes a flight control terminal, an ,intelligent fault-tolerant control algorithm running terminal, and the ground platform includes a ,Mission Planner ground station and a fault-tolerant ground station.

[0057] The fixed-wing UAV aircraft model is used to carry a flight control board, an airborne industrial computer and an airborne data transmission radio, and has good aerodynamics and flight stability;

[0058] The data transmission radio is divided into an airborne data transmission radio and a ground data transmission radio. The airborne data transmission radio is used to obtain the drone status information in the flight control board, send debugging parameters to the flight control board, and communicate with the ground data transmission radio. The ground data transmission radio is used to provide the Mission Planner ground station with drone status information, obtain the parameters to be uploaded by the Mission Planner ground station and the fault-tolerant ground station, and communicate with the airborne data transmission radio. The data transmission radios communicate with each other through the Serial port.

[0059] The flight control board is used to measure the state information of the UAV during flight and control the flight of the UAV. The UAV state information includes attitude angle, attitude angular velocity and flight altitude, etc., and converts the control law into the PWM value of the steering gear, outputs the PWM value to the steering gear, and controls the flight of the UAV. The flight control board can transmit data with the airborne data transmission radio 1 and the airborne industrial computer through the data line;

[0060] The onboard industrial computer is used to carry an intelligent fault-tolerant control algorithm, obtain the drone status information in the flight control board, calculate the fault-tolerant control law, output the fault-tolerant control law, and transmit data with the flight control board through a data line;

[0061] The Mission Planner ground station is used to display the UAV flight information and UAV flight parameter adjustment. It obtains the UAV status information through the digital radio and displays it in the form of numbers and curves, which can more conveniently and intuitively observe the UAV flight status and adjust the flight parameters;

[0062] The fault-tolerant ground station is used to display the fault-tolerant control law, control parameters and intelligent fault-tolerant control algorithm adjustment parameters output by the intelligent fault-tolerant control algorithm. In conjunction with the Mission Planner ground station, the control effect of the intelligent fault-tolerant control algorithm can be more intuitively displayed and the parameters of the intelligent fault-tolerant control algorithm can be adjusted accordingly.

[0063] In this embodiment, the fixed-wing UAV model has a wingspan of 2.8 meters, an overall fuselage length of 2.3 meters, an empty weight of 12 kilograms, a maximum take-off weight of 25 kilograms, a flight altitude of 100-500 meters, and a flight speed of 8-30 meters per second. The fixed-wing UAV model can be equipped with a flight control board, an airborne industrial computer, an airborne digital radio station, and other sensors. The fixed-wing UAV model can realize the injection of stuck faults and loose floating faults. The stuck fault is realized by giving a fixed value to the PWM signal of the servo on a certain rudder surface by the flight control board. The loose floating fault is realized by the loose floating mechanism on the fuselage of the fixed-wing UAV model. The servo can tighten or relax the thin wire used to control the deflection of the rudder surface. When the thin wire is tightened, the rudder surface can be controlled normally. When the thin wire is relaxed, the rudder surface cannot be controlled normally, thereby achieving the effect of loose floating.

[0064] The digital radio is suitable for outdoor flight tests and data transmission needs. The serial port communication part uses the Serial serial port library in Python for data interaction to ensure efficient data transmission and reliability. The airborne digital radio 1 is responsible for two-way data transmission with the flight control board, obtaining the drone status information in the flight control board, and sending flight debugging parameters to the flight control board, including key flight data such as heading, attitude, speed, and flight altitude. At the same time, the airborne digital radio 1 also undertakes the communication task with the ground digital radio 1, transmitting the drone status information to the ground station so that the flight personnel can monitor the flight status of the drone in real time and adjust it as needed. The airborne digital radio 2 is responsible for two-way data transmission with the airborne industrial computer, obtaining the intelligent fault-tolerant control data in the airborne industrial computer, and sending the intelligent fault-tolerant control debugging parameters to the airborne industrial computer. At the same time, the airborne digital radio 2 also undertakes the communication task with the ground digital radio 2. Ground data transmission radios 1 and 2 are used to provide UAV status information to the Mission Planner ground station and intelligent fault-tolerant control data to the fault-tolerant ground station, obtain parameters to be uploaded by the Mission Planner ground station and the fault-tolerant ground station, and communicate with the airborne data transmission radios 1 and 2 respectively.

[0065] The flight control board has built-in high-precision IMU (inertial measurement unit), barometer, GPS module and auxiliary sensors, which can collect and process the attitude, position, speed and other flight data of the drone in real time to ensure accurate control during flight. There is an SD card in the flight control board, which can store the attitude, position, speed and other flight data of the drone during flight. The flight control board is equipped with a flight control program responsible for the normal flight of the drone in the absence of faults. When the pilot manually injects a fault, the flight control board receives the fault information, causing the drone to have a corresponding fault, and immediately switches the normal flight control program to the fault-tolerant flight control program in the onboard industrial computer.

[0066] The onboard industrial computer operating system is Windows 10, which is used to run the intelligent fault-tolerant control algorithm. The fault-tolerant control algorithm can be replaced according to needs and the initial parameters of the fault-tolerant control algorithm can be set. The onboard industrial computer can also store the relevant data of the intelligent fault-tolerant control algorithm and the data of the SD card in the backup flight control board.

[0067] The Mission Planner ground station is used for the initial parameter setting, servo neutral position calibration and flight route setting before the drone takes off. Before the drone takes off, the drone is shaken back and forth and left and right in the stabilization mode to ensure that the servo is near the neutral position and the rudder is deflected correctly when the drone is level. By marking waypoints on the Mission Planner map, the drone can achieve level flight, ascent, descent and turning. The waypoint height is 100-500 meters. The Mission Planner ground station can also be used to display flight data during the drone flight, including real-time digital display and curve graph display.

[0068] The fault-tolerant ground station is developed based on Python's Pyqt library, and can display the digital display and curve display of fault-tolerant control laws and intelligent fault-tolerant control related data in real time, and upload intelligent fault-tolerant control related parameters.

[0069] Embodiment 2:

[0070] Based on the outdoor flight test verification platform provided in Example 1, this embodiment provides a verification method thereof, which is specifically implemented by the following steps:

[0071] 1. The design flow chart of the outdoor test flight verification platform for the intelligent fault-tolerant flight control system of fixed-wing UAV is as follows: Figure 2 As shown. First, the communication protocol between the flight control board, the onboard industrial computer, the Mission Planner ground station and the fault-tolerant ground station is constructed to clarify the data content that each module needs to receive and send; before takeoff, the Mission Planner ground station is used to calibrate the drone servo, initialize the drone flight parameters and write the route, and the intelligent fault-tolerant control algorithm parameters are initialized through the fault-tolerant ground station; during the flight process without faults, the flight control board measures the drone status information, calculates the normal control law, converts the control law into PWM values ​​and outputs it to each servo, thereby controlling the drone to fly autonomously according to the route; after a fault occurs, the flight control board receives the fault-tolerant control law sent by the onboard industrial computer, replaces the original normal control law, and converts it into PWM values ​​and outputs it to each servo, thereby realizing the intelligent fault-tolerant control of the drone under fault conditions and autonomously flying according to the route; the ground station is used to monitor the drone status information and the data related to the intelligent fault-tolerant control algorithm in real time.

[0072] 2. Build intelligent fault-tolerant control algorithms in airborne industrial computers. Figure 3 The terminal flow chart for running the intelligent fault-tolerant control algorithm. The specific design steps are as follows:

[0073] (1) According to the affine mathematical model of the single-channel attitude loop of the UAV and the fault mechanism, the fault model is established and written into the airborne industrial control computer. This patent considers the rudder stuck fault, rudder loose fault, rudder gap fault and rudder defect fault, and the system is decoupled into three channels. Considering the affine mathematical model of the single-channel attitude loop of the UAV under fault conditions, it is:

[0074] (1);

[0075] in, and are attitude angle and attitude angular velocity respectively, and is the aerodynamic parameter, is the control efficiency factor, is the control signal deviation caused by the fault, is the unknown disturbance, is the control law. When there is no fault , stuck fault , loose and floating fault , gap fault , defect fault .

[0076] (2) Derive the fault-tolerant control law and write it into the intelligent fault-tolerant control algorithm running terminal. As fault-related items, according to the UAV fault model, a radial basis function neural network (RBFNN) is designed to fit and estimate the fault-related items and design the control law.

[0077] Sliding mold (2);

[0078] in, is the attitude angle tracking deviation, is the fractional order, D is a fractional calculus operator, is the symbolic function, is the desired attitude angle.

[0079] By taking the derivative of (2), we can get:

[0080] (3);

[0081] Selecting the exponential reaching law

[0082] (4);

[0083] in, are the parameters to be designed.

[0084] The control law is designed as follows:

[0085] (5);

[0086] in, for The estimated value of .

[0087] Design RBF neural network for fault related items Perform fitting and design the following RBF neural network:

[0088] (6);

[0089] in, is the estimated value of the ideal network weight, is the Gaussian kernel function of the network, which is as follows:

[0090] (7);

[0091] in, is the input of the network, and are the parameters to be designed.

[0092] In order to overcome the input of RBF neural network The algebraic loop problem caused by this is firstly Passed to Butterworth low-pass filter, we get ,Right now

[0093] (8);

[0094] in, is the filtering error.

[0095] (3) The control system of the drone consists of two core components: the outer loop position control and the inner loop attitude control. In a fault-free state, both the outer loop and the inner loop are implemented by the normal flight control program in the flight control board. However, once the drone encounters a fault, the control strategy will switch: the outer loop continues to be maintained by the normal flight control program in the flight control board, while the inner loop is taken over by the intelligent fault-tolerant control flight control program in the onboard industrial computer to ensure the stability and safety of the drone.

[0096] 3. Build the flight control terminal, Figure 4 The following is the flow chart of the flight control terminal. The flight control terminal is implemented by the normal flight control program in the flight control board, using the open source APM (ArduPilot Mega) flight control.

[0097] (1) Modify the firmware. This platform uses an X-shaped tail rudder fixed-wing drone. The aileron controls the roll, and the tail controls the pitch and yaw. Figure 5As shown, the coordinate system is the Soviet coordinate system, the positive direction of the X-axis points forward along the body, the positive direction of the Y-axis points vertically upward, and the positive direction of the Z-axis points horizontally to the right according to the right-hand system. Rudders 1, 2, 3, and 4 are respectively the upper right, upper left, lower left, and lower right of the X-type tail rudder. The positive and negative rudder deflection angles are defined as, looking forward along the tail, the rudder surface edge deflects upward in a positive direction. The rudder deflection angles of rudders 1, 2, 3, and 4 are defined as , , , The roll control output is , the yaw control output is , the pitch control output is Then the control distribution array of the tail wing can be designed as:

[0098] (9);

[0099] Generate new firmware after modifying the mixer.

[0100] (2) Burn the firmware. Burn the newly generated firmware to the flight control board through the Mission Planner ground station.

[0101] (3) Test the firmware. First, switch the flight mode to manual mode, and move the remote control stick to test whether the rudder deflection is normal; then switch the flight mode to stabilization mode, and move the remote control stick to test whether the rudder deflection is normal. Then lift the drone and shake it back and forth and left and right to test whether the rudder deflection is normal.

[0102] 4. Build the Mission Planner ground station and the fault-tolerant ground station. The communication links of the two ground stations are independent and do not interfere with each other. The Mission Planner ground station uses the official communication method and does not require additional configuration. After pairing the airborne data radio 1 and the ground data radio 1, the drone status information can be displayed on the Mission Planner ground station page and the drone parameters can be uploaded;

[0103] The fault-tolerant ground station uses Serial communication to clarify the data, data type and byte number transmitted between each module and determine the communication protocol. Custom read and write functions are used to realize data transmission between airborne data transmission radio 2 and ground data transmission radio 2. The display interface of the fault-tolerant ground station is implemented using PyQt, which can display fault-tolerant control related data in real time, and use the oscilloscope module to display the data change curve in real time.

[0104] 5. In order to better understand the outdoor test flight verification platform of the above-mentioned fixed-wing UAV intelligent fault-tolerant flight control system, the effectiveness of the proposed platform is demonstrated through specific examples below.

[0105] Route settings such as Figure 6As shown, the altitude of waypoint ③ is 200 meters, the altitude of other waypoints is 150 meters, the horizontal distance between adjacent waypoints is 300 meters, the UAV flies counterclockwise according to the set route, and the No. 1 rudder stuck fault is injected at waypoint ②. Under the condition of No. 1 rudder stuck fault and intelligent fault-tolerant control intervention, the UAV completes the ascent, descent, turning and level flight actions in sequence, cuts out the fault at waypoint ⑦, and completes a circle of flight test.

[0106] During the flight of the drone, it can be seen in the air and on the Mission Planner ground station that the drone has lowered or raised its head at the moment of fault injection. After the intelligent fault-tolerant control algorithm intervenes, the drone's attitude quickly returns to a balanced state and can complete ascent, descent, turning and level flight according to the set route. After a fault occurs, the fault-tolerant algorithm can estimate the value of the fault-related items and compensate for them in the control law.

[0107] In summary, this example proves that the above test flight system can effectively implement the intelligent fault-tolerant control algorithm of fixed-wing UAVs, and researchers can intuitively understand the flight status of UAVs in real environments and the effect of fault-tolerant algorithms. At the same time, the platform can modify the firmware for different fixed-wings, and carry different fault-tolerant control algorithms in the onboard industrial computer. Both the UAV and the algorithm are modifiable, which provides a good experimental basis for outdoor test flights of other UAV fault-tolerant flight control systems and saves costs.

[0108] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. An outdoor test flight verification platform for intelligent fault-tolerant flight control system of fixed-wing UAV, characterized in that: The platform includes: a fixed-wing UAV aircraft model, a digital radio, a flight control board, an airborne industrial computer, a Mission Planner ground station and a fault-tolerant ground station; wherein, The fixed-wing UAV aircraft model is used to carry a flight control board, an airborne industrial computer, an airborne data transmission radio and sensor equipment; The data transmission radio is divided into an airborne data transmission radio and a ground data transmission radio. The airborne data transmission radio is used to obtain the drone status information in the flight control board, send debugging parameters to the flight control board, and communicate with the ground data transmission radio. The ground data transmission radio is used to provide the Mission Planner ground station with drone status information, obtain the parameters to be uploaded by the Mission Planner ground station and the fault-tolerant ground station, and communicate with the airborne data transmission radio. The data transmission radios communicate with each other through the Serial port. The flight control board is used to measure the state information of the UAV during flight and control the flight of the UAV, convert the control law into the PWM value of the steering gear, output the PWM value to the steering gear, and control the flight of the UAV; the flight control board exchanges information with the airborne data transmission radio and the airborne industrial computer; The onboard industrial computer is used to carry an intelligent fault-tolerant control algorithm, obtain the drone status information in the flight control board, calculate the fault-tolerant control law, output the fault-tolerant control law, and send it to the flight control board; the intelligent fault-tolerant control algorithm is specifically: S1 establishes an affine mathematical model of a single-channel attitude loop of a UAV under fault consideration, wherein the attitude loop affine mathematical model includes a UAV fault model; S2 is based on the UAV fault model and designs a radial basis function neural network (RBFNN) to fit and estimate fault-related items and design control laws; The S2 is specifically: Set the sliding mode: （2）； in, is the attitude angle tracking deviation, is the attitude angle, is the fractional order, D is a fractional calculus operator, is a symbolic function; is the desired attitude angle; By taking the derivative of (2), we can obtain: （3）； in, and is the aerodynamic parameter; Choose an exponential reaching law: （4）； in, is the parameter to be designed; The control law is designed as follows: （5）； in, for An estimated value of is the control law; is the fault related item; Design RBF neural network for fault related items Perform fitting and design the following RBF neural network: （6）； in, is the estimated value of the ideal network weight, is the Gaussian kernel function of the network, which is as follows: （7）； in, is the input of the network, and are the parameters to be designed; The control law Passed to the Butterworth low-pass filter, the filtered control law is obtained ,Right now （8）； in, is the filtering error; The Mission Planner ground station is used to display the UAV flight information and UAV flight parameters, and obtain the UAV status information through the data transmission radio; The fault-tolerant ground station is used to display the fault-tolerant control law, control parameters and intelligent fault-tolerant control algorithm adjustment parameters output by the intelligent fault-tolerant control algorithm, and cooperate with the Mission Planner ground station to adjust the intelligent fault-tolerant control algorithm parameters.

2. According to claim 1, the outdoor test flight verification platform for the intelligent fault-tolerant flight control system of a fixed-wing UAV is characterized in that: The drone status information includes attitude angle, attitude angular velocity and flight altitude.

3. The outdoor test flight verification platform for the intelligent fault-tolerant flight control system of a fixed-wing UAV according to claim 1 is characterized in that: The digital radio is used for outdoor flight tests and data transmission. The serial port part uses the Serial serial port library in Python for data interaction. The airborne data transmission radio is used for two-way data transmission with the flight control board, obtaining the UAV status information in the flight control board, and sending flight debugging parameters to the flight control board; the airborne data transmission radio is also used for communication with the ground data transmission radio, transmitting the UAV status information to the Mission Planner ground station for adjustment according to the designed parameters; the airborne data transmission radio is also used for two-way interaction with the airborne industrial computer, obtaining the intelligent fault-tolerant control data in the airborne industrial computer, and sending the intelligent fault-tolerant control debugging parameters to the airborne industrial computer.

4. The outdoor test flight verification platform for the intelligent fault-tolerant flight control system of a fixed-wing UAV according to claim 1 is characterized in that: The flight control board has built-in high-precision inertial measurement unit IMU, barometer, GPS module and auxiliary sensors, which are used to collect and process the flight data of the drone in real time; the flight control board is equipped with a flight control program for the normal flight of the drone under fault-free conditions. When the pilot manually injects a fault, the flight control board receives the fault information, causing the drone to generate a corresponding fault, and immediately switches the normal flight control program to the fault-tolerant flight control program in the onboard industrial computer.

5. The outdoor test flight verification platform for the intelligent fault-tolerant flight control system of a fixed-wing UAV according to claim 1 is characterized in that: The Mission Planner ground station is used for initial parameter setting, servo neutral position calibration and flight route setting before the drone takes off. The drone is shaken in stabilization mode before takeoff, and waypoints are marked on the Mission Planner map to achieve level flight, ascent, descent and turning of the drone. The Mission Planner ground station is used for displaying flight data during the flight of the drone.

6. The outdoor test flight verification platform for the intelligent fault-tolerant flight control system of a fixed-wing UAV according to claim 1 is characterized in that: The fault-tolerant ground station is developed based on Python's Pyqt library, and displays the digital display and curve display of fault-tolerant control laws and intelligent fault-tolerant control related data in real time, and uploads intelligent fault-tolerant control related parameters.

7. The outdoor test flight verification platform for the intelligent fault-tolerant flight control system of a fixed-wing UAV according to claim 1 is characterized in that: The attitude ring affine mathematical model is: （1）； in, is the attitude angular velocity, is the control efficiency factor, is the control signal deviation caused by the fault, is an unknown disturbance; Will as a fault related item.

Images