Flight test system for realizing online verification and debugging of fault diagnosis algorithm

Abstract

The invention discloses a flight test system for realizing online verification and debugging of a fault diagnosis algorithm, and relates to the field of aircraft fault monitoring. The flight test system selects a fixed-wing unmanned aerial vehicle and a secondarily developed flight control algorithm according to a typical fault generation mechanism of an execution mechanism of the fixed-wing unmanned aerial vehicle; the method comprises the following steps of: simulating typical faults of two actuating mechanisms, namely control surface jamming and control surface loosening and floating, and performing online real-time monitoring on multi-dimensional flight state information of the fixed-wing unmanned aerial vehicle by combining an airborne computer carrying an online fault diagnosis algorithm and utilizing flight state data sent by a flight controller in real time; in cooperation with an independently developed fault diagnosis information monitoring and parameter adjusting ground station, online verification and debugging can be carried out on a fault diagnosis algorithm aiming at the fault of the executing mechanism of the fixed-wing unmanned aerial vehicle.

Description

Technical Field

[0001] The invention relates to the field of aircraft fault monitoring, and in particular to a flight test system for realizing online verification and debugging of a fault diagnosis algorithm. Background Art

[0002] As the functions of modern aircraft continue to expand, their structures have become more complex. At the same time, they are affected by a variety of external factors, and the probability of failure is gradually increasing, and their reliability is facing challenges. If an aircraft fails to be discovered and handled in time during a mission, it will not only affect the completion of the mission, but may also cause the aircraft to be abandoned, thereby reducing the possibility of its reuse. Therefore, simulating and diagnosing typical aircraft faults is of great significance for verifying aircraft fault diagnosis algorithms, etc. It is a necessary guarantee to ensure the normal operation of the aircraft and the basis for further taking effective fault-tolerant measures.

[0003] Considering that the existing fault simulation technology mainly conducts targeted modeling in simulation software for the components, systems and complete machines that need to be simulated according to specific simulation requirements, and simulates the faults through the established models. However, considering that even the most complex environmental modeling is still difficult to approximate the impact of the actual environment on the flight status of the aircraft, the fault diagnosis algorithm debugged by the flight status data obtained in this way will show problems such as poor robustness in actual applications. Summary of the invention

[0004] In view of the above problems and technical requirements, the inventors have proposed a flight test system for realizing online verification and debugging of fault diagnosis algorithms. The technical solution of the present invention is as follows:

[0005] A flight test system for realizing online verification and debugging of a fault diagnosis algorithm comprises the following steps:

[0006] drone and sensor set;

[0007] A fault simulation component, including a remote controller and a loose floating mechanism installed on the UAV, wherein the remote controller is used to provide a fault simulation signal;

[0008] The flight controller is connected to the sensor group, the remote controller and the rudder servo, and is used to control the corresponding rudder of the drone to simulate the corresponding fault according to the fault simulation signal;

[0009] The onboard computer provides an operating framework for the fault diagnosis algorithm and receives flight status data from the flight controller to perform online fault diagnosis tests.

[0010] The flight control ground station establishes a communication connection with the flight controller through a data transmission radio to plan the automatic route of the drone and adjust the relevant parameters of the flight controller;

[0011] The fault diagnosis information monitoring and parameter adjustment ground station establishes a communication connection with the onboard computer through a data transmission radio, which is used to monitor the flight status and fault diagnosis information of the UAV, and realize online debugging of parameters related to the fault diagnosis algorithm.

[0012] Its further technical solution is that the UAV adopts a fixed-wing UAV, including propellers, fuselage, ailerons, and X-tail, and a battery compartment, flight controller and onboard computer are installed inside the fuselage;

[0013] The propeller is attached to the front of the fuselage to provide thrust;

[0014] Ailerons are located on both sides of the fuselage and are used to provide rolling moment;

[0015] The X-tail is used to provide pitch and yaw moments and serves as the object of fault simulation for the drone.

[0016] A further technical solution is that the floating mechanism includes a base, a rotating component, a connecting rod, and two connecting wires of the same length;

[0017] The base is used to fix the entire loose floating mechanism to the UAV fuselage;

[0018] The rotating shaft of the rotating component is coaxially connected to a steering gear and fixed on the base. The steering gear, driven by the flight controller, drives the rotating shaft to drive the rotating surface thereon to rotate in clockwise and counterclockwise directions.

[0019] The connecting rod is connected to a servo, and one end of the connecting rod is fixed to the base, and the other end is regarded as the movable end;

[0020] One end of the first connecting line is fixed to the movable end, and the other end passes through a small hole on the first end of the rotating surface and is connected to one side of a movable part of a rudder surface; one end of the second connecting line is fixed to the movable end, and the other end passes through a small hole on the second end of the rotating surface and is connected to the other side of the movable part of the same rudder surface; wherein the first end and the second end of the rotating surface are opposite, and when the rotating surface is in the initial position, the connecting line between the first end and the second end is perpendicular to the fuselage;

[0021] Driven by the flight controller, the servo moves the movable end along the direction of the fuselage to drive the two connecting lines to be in a tight or loose state; when the rotating surface rotates in different directions, the movable part of the rudder surface connected to the connecting line is driven to steer in the corresponding direction.

[0022] A further technical solution is to control the corresponding control surface of the UAV to simulate the corresponding fault according to the fault simulation signal, including:

[0023] The remote controller is provided with a fault mode switch, a fault type lever, and a fault position lever, and generates a corresponding fault simulation signal based on the selected switch and lever position;

[0024] When the flight controller analyzes the fault simulation signal and finds that there is no fault, the flight controller outputs the rudder servo signal according to the original flight control parameters to control the drone to fly normally;

[0025] When the flight controller interprets the fault simulation signal as a fault, the flight controller outputs corresponding rudder servo signals according to the simulated fault type and fault location, and simulates a rudder stuck fault or a rudder loose floating fault through the loose floating mechanism.

[0026] Its further technical solution is that, when the fault type is a rudder stuck fault, the flight controller reads the real-time control signal of the selected faulty rudder servo and continuously outputs it to the faulty rudder servo as a constant control signal, so that the rotating surface of the loose floating mechanism is fixed at the current position. At this time, the faulty rudder is stuck at the immediate angle and no longer responds to the original flight control parameters.

[0027] Its further technical solution is that when the fault type is a loose and floating rudder fault, the flight controller sends a minimum control signal to the selected faulty rudder servo, so that the connecting line of the loose and floating mechanism is in a loose state, and the faulty rudder is no longer controlled by the servo.

[0028] Its further technical solution is that the working process of the flight test system includes:

[0029] After initializing the flight test system, the flight controller controls the UAV to automatically fly along the preset route;

[0030] Select the type and location of the simulated fault on the remote controller, generate a fault simulation signal and send it to the flight controller. The flight controller controls the corresponding control surface through the floating mechanism to simulate the corresponding fault according to the received fault simulation signal.

[0031] The flight controller transmits real-time flight status data to the onboard computer equipped with the fault diagnosis algorithm to obtain fault diagnosis information;

[0032] The fault diagnosis information monitoring and parameter adjustment ground station displays the fault diagnosis information and flight status data through data communication. In this process, the fault diagnosis information monitoring and parameter adjustment ground station provides a parameter adjustment interface to realize online debugging of relevant parameters of the fault diagnosis algorithm.

[0033] A further technical solution is to initialize the flight test system, including:

[0034] Re-power on the flight controller, drone, onboard computer, and data radio in order;

[0035] Restart the data path and fault diagnosis algorithm of the onboard computer, and restart the fault diagnosis information monitoring and parameter adjustment ground station and the flight control ground station.

[0036] Its further technical solution is that the working process of the flight test system also includes:

[0037] According to different fault diagnosis algorithms, the information displayed and monitored by the fault diagnosis information monitoring and parameter adjustment ground station and the algorithm parameters that need to be debugged are replaced, and the flight status data needed by the fault diagnosis algorithm carried by the onboard computer is replaced to ensure the versatility of the flight test system.

[0038] Its further technical solution is that the fault diagnosis information monitoring and parameter adjustment ground station is a visual interface developed based on Python.

[0039] The beneficial technical effects of the present invention are:

[0040] The above-mentioned flight test system analyzes the typical fault generation mechanism of the fixed-wing UAV actuator, constructs a fault simulation component that realizes two typical faults, and combines the secondary developed flight control algorithm to simulate the two typical faults of the actuator's rudder sticking and rudder loosening at any time. In conjunction with the fault diagnosis algorithm installed on the onboard computer, the UAV can be directly debugged in the actual engineering application stage, and the online debugging of the fault diagnosis algorithm can be realized through the coordinated cooperation of the independently developed ground station, so that the fault diagnosis algorithm truly has engineering application value, rather than only being able to be used on the simulation model.

[0041] Before the flight test system is used, all components must be initialized on the ground and the pilot must remotely control the drone to take off and enter the automatic route. After actual testing, the flight test system's command sending, fault simulation device response, data acquisition and communication rate are greater than 100Hz, and it can support ground observation and debugging of fault diagnosis algorithms. It can be promoted and applied to the online debugging and verification of fault diagnosis algorithms for various fixed-wing drones, improving the robustness of fault diagnosis algorithms and having broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is an architectural diagram of a flight test system for implementing online verification and debugging of fault diagnosis algorithms provided in this application.

[0043] Figure 2 These are structural diagrams of the loose floating mechanism provided in this application from different perspectives, wherein (a) is the main view and (b) is the front view.

[0044] Figure 3 It is a workflow diagram of the flight test system provided by this application. DETAILED DESCRIPTION

[0045] The specific implementation of the present invention will be further described below in conjunction with the accompanying drawings.

[0046] An embodiment of the present application provides a flight test system for realizing online verification and debugging of a fault diagnosis algorithm. Please refer to Figure 1 As shown, the flight test system includes a UAV, a flight controller, a sensor group, a fault simulation component, an onboard computer, a flight control ground station and a fault diagnosis information monitoring and parameter adjustment ground station.

[0047] In this embodiment, the UAV is a fixed-wing UAV, including a propeller 1, a fuselage 2, ailerons 3, and an X-tail 4. The propeller 1 is connected to the front of the fuselage to provide pulling force. A battery compartment, a flight controller, and an onboard computer are installed inside the fuselage 2; optionally, the flight controller is fixed at the center of gravity of the fixed-wing UAV. The ailerons 3 are located on both sides of the fuselage to provide a rolling moment. The X-tail 4 is used to provide pitch and yaw moments, and at the same time has a certain fault tolerance capability, which can improve the safety of the flight test. In this embodiment, the X-tail 4 serves as a fault simulation object of the UAV. The sensor group includes an inertial navigation module, a GPS, an airspeed meter, etc.

[0048] The fault simulation component includes a remote controller, a fault simulation program, and a floating mechanism installed on the drone. The remote controller is mainly used to remotely control the flight of the fixed-wing drone. It is also provided with a fault mode switch, a fault type lever, and a fault position lever. The fault mode switch is used to select whether the drone performs fault simulation. The fault type lever is used to select a rudder stuck fault or a rudder floating fault. The fault position lever is used to select a tail wing in the X-tail 4 as a faulty rudder. Based on the selected switch and lever position, a corresponding fault simulation signal is generated and sent to the flight controller. The fault simulation program is mounted on the flight controller, and the fault simulation process of controlling the X-tail is performed based on the fault simulation program.

[0049] Each tail wing in the X-Tail 4 is equipped with a loose floating mechanism to simulate the corresponding failure of the control surface. Figure 2As shown in (a) and (b), the floating mechanism includes a base 5, a rotating component 6, a connecting rod 7, and two connecting wires of the same length. The base 5 is used to fix the entire floating mechanism on the UAV fuselage 2. The rotating shaft of the rotating component 6 is coaxially connected to a steering gear and fixed on the base 5. The steering gear drives the rotating shaft to drive the rotating surface thereon to rotate in the clockwise and counterclockwise directions under the drive of the flight controller. The connecting rod 7 is connected to a steering gear, and one end of the connecting rod is fixed to the base, and the other end is regarded as a movable end. The steering gear drives the movable end to move along the fuselage direction under the drive of the flight controller. One end of the first connecting wire 8a is fixed to the movable end, and the other end passes through the small hole on the first end of the rotating surface and is connected to one side of a movable part of a rudder surface. Similarly, one end of the second connecting wire 8b is fixed to the movable end, and the other end passes through the small hole on the second end of the rotating surface and is connected to the other side of the same movable part of the rudder surface. Among them, the first end and the second end of the rotating surface are opposite, and when the rotating surface is in the initial position shown in the figure, the connecting line between the first end and the second end is perpendicular to the fuselage. The connection point of the connection lines 8a, 8b at the movable part of the control surface is at the same distance from the tail of the stabilizer surface of the tail wing.

[0050] According to the connection principle of the above components, the movement of the connecting rod 7 can drive the two connecting lines to be in a tight or loose state, and when the rotating surface rotates in different directions in the tight state, it can drive the movable part of the rudder surface connected to the connecting line to steer in the corresponding direction. Figure 2 As shown, when the rotating surface rotates counterclockwise, the first connecting line 8a is pulled forward so that the tail connected thereto is steered upward; when the rotating surface rotates clockwise, the second connecting line 8b is pulled forward so that the tail connected thereto is steered downward.

[0051] The flight controller connects the sensor group, remote control and rudder servos. It is equipped with a secondary developed flight control algorithm to control the automatic flight of the fixed-wing UAV and provide flight status data to the onboard computer. The flight controller also controls the corresponding rudder of the UAV to simulate the corresponding fault according to the fault simulation signal provided by the remote control, including:

[0052] (1) When the flight controller interprets the fault simulation signal as fault-free, the flight controller outputs the control surface and servo signals according to the original flight control parameters to control the normal flight of the UAV. That is, the flight controller applies the maximum control signal to each control surface and servo, so that the connecting rod moves toward the tail to a preset position. At this time, the two connecting lines are in a tight state, so that the corresponding control surface and servo are closely connected, and the X-tail is controlled to achieve normal output signal response.

[0053] (2) When the flight controller interprets the fault simulation signal as a fault, the flight controller outputs the corresponding rudder servo signal according to the simulated fault type and fault location, and simulates a rudder stuck fault or a rudder loose floating fault through the loose floating mechanism.

[0054] A stuck fault is a fault type in which the control surface is instantly stuck at a certain angle due to a stuck fault on the basis of normal flight without faults. Therefore, when the actuator fault demand indicates that the simulated fault type is a stuck fault, the three levers of the remote controller fault mode, fault type and fault position are switched to the corresponding positions, thereby sending a stuck fault simulation signal to the flight controller. After receiving the signal, the flight controller reads the instant control signal of the selected faulty control surface servo, and continuously outputs it to the faulty control surface servo as a constant control signal, so that the rotating surface of the loose floating mechanism is fixed at the current position. At this time, the faulty control surface is stuck at the instant angle and no longer responds to the original flight control parameters.

[0055] The loose floating fault is a fault condition in which the output shaft of the servo is momentarily unloaded due to the loose floating of the rudder surface on the basis of normal flight without faults. Therefore, when the fault type simulated by the actuator fault demand indication is a loose floating fault, the three levers of the remote controller fault mode, fault type and fault position are switched to the corresponding positions, thereby sending a loose floating fault simulation signal to the flight controller. After receiving the signal, the flight controller sends a minimum control signal to the selected faulty rudder surface servo, so that the connection line of the loose floating mechanism is in a loose state. At this time, the faulty rudder surface is no longer controlled by the servo, that is, when the rotating surface rotates counterclockwise, the first connecting line 8a cannot achieve the forward pulling effect due to looseness, so that the tail connected to it cannot steer upward and cannot respond to control; when the rotating component rotates clockwise, the second connecting line 8b cannot achieve the forward pulling effect due to looseness, so that the tail connected to it cannot steer downward and cannot respond to control.

[0056] The onboard computer provides a running framework for the fault diagnosis algorithm, and also provides serial communication function to receive the flight status data provided by the flight controller and save it in the global variable library, providing offline data storage and online data support for the fault diagnosis algorithm. Optionally, the onboard computer uses the Windows system and supports multiple programming environments such as Python and C++.

[0057] The flight control ground station establishes a communication connection with the flight controller through communication equipment (such as data transmission radio) to adjust the relevant parameters of the flight controller (such as the neutral value of the control surface, PID coefficient, etc.), and supports changing the flight status and automatic route according to the needs of fault simulation. The fault diagnosis information monitoring and parameter adjustment ground station is a visualization interface developed based on Python. It establishes a communication connection with the onboard computer through the data transmission radio to realize the visualization of flight status and fault diagnosis information. At the same time, it provides a parameter adjustment interface to realize the online debugging of the relevant parameters of the fault diagnosis algorithm.

[0058] Based on the above flight test system, typical faults of actuators can be simulated and used for online verification and debugging of fault diagnosis algorithms. Please refer to the working process of the flight test system. Figure 2The flowchart shown includes the following contents:

[0059] After initializing the flight test system, the flight controller controls the drone to automatically fly along the preset route, and sends the flight status data to the onboard computer through the serial port. The onboard computer saves the received flight status data and writes it into the global variable library. Select the type and location of the simulated fault on the remote control, form a fault simulation signal and send it to the flight controller. The flight controller controls the corresponding control surface through the loose floating mechanism to simulate the corresponding fault according to the received fault simulation signal, and transmits the real-time flight status data to the onboard computer equipped with the fault diagnosis algorithm. The fault diagnosis algorithm can read the corresponding flight status data as needed, realize the online test of fault diagnosis, and obtain fault diagnosis information.

[0060] After data communication, the fault diagnosis information monitoring and parameter adjustment ground station displays the fault diagnosis information and flight status data. In this process, the data communication between the onboard computer and the fault diagnosis information monitoring and parameter adjustment ground station allows the diagnosis results and key information of each round of the fault diagnosis algorithm to be written into the global variable library, so as to be transmitted to the ground and displayed on the fault diagnosis information monitoring and parameter adjustment ground station, so as to realize the real-time monitoring of the conclusions and key information of the fault diagnosis algorithm. At the same time, the data communication between the onboard computer and the fault diagnosis information monitoring and parameter adjustment ground station allows the fault diagnosis algorithm to create a parameter adjustment interface, write the debugging parameters through the fault diagnosis information monitoring and parameter adjustment ground station and upload them to the onboard computer, and the onboard computer receives and writes them into the global variable library for the fault diagnosis algorithm on board, so as to realize the online debugging of the fault diagnosis algorithm.

[0061] Optionally, according to different fault diagnosis algorithms, the information displayed and monitored by the fault diagnosis information monitoring and parameter adjustment ground station and the algorithm parameters that need to be debugged can be replaced, and the flight status data required by the fault diagnosis algorithm carried by the onboard computer can be replaced to ensure the versatility of the flight test system. For the fault simulation components that have simulated the actuator failure, the fault can be cut out at any time to ensure the safety of the flight test system.

[0062] In the above work process, the flight test system needs to be initialized first, including:

[0063] (1) Re-power on the flight controller, drone, onboard computer, and data transmission radio in order. Specifically, first re-power on the flight controller to reset the attitude, position and other measurement information and the servo output signal. Then re-power on the fixed-wing drone so that the control voltage of the propeller and servo returns to the neutral value. After the above operations are completed, re-power on the onboard computer and data transmission radio and start them. After the above process, the initialization operation of the fixed-wing drone, flight controller, onboard computer, and data transmission radio is completed.

[0064] (2) Restart the data path and fault diagnosis algorithm of the onboard computer, and restart the fault diagnosis information monitoring and parameter adjustment ground station and the flight control ground station.

[0065] In addition, the flight test system can also add a variety of sensors according to the needs of different fault diagnosis algorithms, so as to expand the measurement of more parameters to support different data acquisition needs; at the same time, the number of flight states can be freely increased or decreased for the data path. The flight test system has good scalability and practicality through the replacement and increase of components, and has broad application prospects for online debugging and verification of fault diagnosis algorithms.

[0066] Taking fault detection and fault identification in the fault diagnosis algorithm as an example, the fault detection and fault identification algorithm is written in the form of an online process and embedded in the algorithm position reserved by the onboard computer. According to the algorithm requirements, the flight controller sends the flight status data such as attitude angle, attitude angular velocity, and control amount of the rudder to the onboard computer through the serial port. The onboard computer decodes the serial port to receive these data and writes them into the global variable library. In each cycle of the algorithm operation, the fault detection and fault identification algorithm reads the data of the global variable library and completes its own operation. In the above main process operation, the flight test system will also transmit the threshold, detection value and final result of the fault detection and fault identification algorithm to the fault diagnosis information monitoring and parameter adjustment ground station for visual display, so as to provide an important basis for real-time debugging of the algorithm; at the same time, the threshold initialization time, threshold amplification ratio, reinitialization and other debugging options preset by the fault diagnosis information monitoring and parameter adjustment ground station can upload the real-time modified parameters to the onboard computer to realize the online debugging of the fault detection and fault identification algorithm.

[0067] It should be noted that the onboard computers, flight controllers and other equipment used in this application all support real-time fault simulation and signal output of no more than 10ms. The support of these high real-time hardware devices enables the flight test system of this application to achieve high real-time control and fault simulation through program control, so that the simulated faults have a higher rate of change, thereby meeting the real-time requirements of online debugging and verification of fault diagnosis algorithms.

[0068] The above is only a preferred embodiment of the present application, and the present invention 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 invention should be considered to be included in the protection scope of the present invention.

Claims

1. A flight test system for realizing online verification and debugging of fault diagnosis algorithms, characterized in that: include: drone and sensor set; A fault simulation component, comprising a remote controller and a loose floating mechanism installed on the UAV, wherein the remote controller is used to provide a fault simulation signal; A flight controller, connected to the sensor group, the remote controller and the rudder servo, for controlling the corresponding rudder of the UAV to simulate the corresponding fault according to the fault simulation signal; The onboard computer provides an operating framework for the fault diagnosis algorithm and receives the flight status data provided by the flight controller to perform online fault diagnosis testing; The flight control ground station establishes a communication connection with the flight controller via a data transmission radio, and is used to plan the automatic route of the UAV and adjust the relevant parameters of the flight controller; The fault diagnosis information monitoring and parameter adjustment ground station establishes a communication connection with the onboard computer through a data transmission radio, and is used to monitor the flight status and fault diagnosis information of the UAV, and to realize online debugging of parameters related to the fault diagnosis algorithm.

2. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 1, characterized in that: The UAV is a fixed-wing UAV, including a propeller, a fuselage, ailerons, and an X-tail, and a battery compartment, the flight controller, and the onboard computer are installed inside the fuselage; The propeller is connected to the front of the fuselage to provide pulling force; The ailerons are located on both sides of the fuselage and are used to provide rolling moment; The X-tail is used to provide pitch and yaw moments and serves as a fault simulation object of the drone.

3. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 1, characterized in that: The floating mechanism comprises a base, a rotating component, a connecting rod, and two connecting wires of the same length; The base is used to fix the entire loose floating mechanism on the UAV fuselage; The rotating shaft of the rotating component is coaxially connected to a steering gear and fixed on the base, and the steering gear drives the rotating shaft to drive the rotating surface thereon to rotate in clockwise and counterclockwise directions under the drive of the flight controller; The connecting rod is connected to a steering gear, and one end of the connecting rod is fixed to the base, and the other end is regarded as a movable end; One end of the first connecting line is fixed to the movable end, and the other end passes through a small hole on the first end of the rotating surface and is connected to one side of a movable part of a rudder surface; one end of the second connecting line is fixed to the movable end, and the other end passes through a small hole on the second end of the rotating surface and is connected to the other side of the movable part of the same rudder surface; wherein the first end and the second end of the rotating surface are opposite, and when the rotating surface is in the initial position, the connecting line between the first end and the second end is perpendicular to the fuselage; The servo drives the flight controller to move the movable end in the direction of the fuselage to drive the two connecting lines to be in a tight or loose state; when the rotating surface rotates in different directions, the movable part of the rudder surface connected to the connecting line is driven to steer in the corresponding direction.

4. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 1, characterized in that: Controlling the corresponding control surface of the UAV to simulate the corresponding fault according to the fault simulation signal includes: The remote controller is provided with a fault mode switch, a fault type lever, and a fault position lever, and generates a corresponding fault simulation signal based on the selected switch and lever position; When the flight controller interprets the fault simulation signal as being fault-free, the flight controller outputs a control surface servo signal according to the original flight control parameters to control the UAV to fly normally; When the flight controller interprets the fault simulation signal as a fault, the flight controller outputs a corresponding rudder servo signal according to the simulated fault type and fault location, and simulates a rudder stuck fault or a rudder loose floating fault through the loose floating mechanism.

5. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 4, characterized in that: When the fault type is a rudder stuck fault, the flight controller reads the real-time control signal of the selected faulty rudder servo and continuously outputs it to the faulty rudder servo as a constant control signal, so that the rotating surface of the loose floating mechanism is fixed at the current position. At this time, the faulty rudder is stuck at the real-time angle and no longer responds to the original flight control parameters.

6. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 4, characterized in that: When the fault type is a loose and floating fault of the rudder surface, the flight controller sends a minimum control signal to the selected faulty rudder surface servo, so that the connecting line of the loose and floating mechanism is in a loose state, and the faulty rudder surface is no longer controlled by the servo.

7. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 1, characterized in that: The working process of the flight test system includes: After the flight test system is initialized, the flight controller controls the UAV to automatically fly along a preset route; Selecting the type and location of the simulated fault on the remote controller, generating a fault simulation signal and sending it to the flight controller, and the flight controller controls the corresponding control surface through the floating mechanism to simulate the corresponding fault according to the received fault simulation signal; The flight controller transmits the real-time flight status data to the onboard computer equipped with the fault diagnosis algorithm to obtain fault diagnosis information; The fault diagnosis information monitoring and parameter adjustment ground station displays the fault diagnosis information and flight status data through data communication; in this process, the fault diagnosis information monitoring and parameter adjustment ground station provides a parameter adjustment interface to realize online debugging of relevant parameters of the fault diagnosis algorithm.

8. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 7, characterized in that: Initializing the flight test system, including: Re-power on the flight controller, the drone, the onboard computer, and the data transmission radio in order; The data path and fault diagnosis algorithm carried by the onboard computer are restarted, and the fault diagnosis information monitoring and parameter adjustment ground station and the flight control ground station are restarted.

9. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 7, characterized in that: The working process of the flight test system also includes: According to different fault diagnosis algorithms, the information displayed and monitored by the fault diagnosis information monitoring and parameter adjustment ground station and the algorithm parameters that need to be debugged are replaced, and the flight status data required for the fault diagnosis algorithm carried by the onboard computer is replaced to ensure the versatility of the flight test system.

10. The flight test system for realizing online verification and debugging of fault diagnosis algorithm according to claim 1, characterized in that: The fault diagnosis information monitoring and parameter adjustment ground station is a visual interface developed based on Python.

Images