An unmanned aerial vehicle servo flap system, a wing of an unmanned aerial vehicle and an unmanned aerial vehicle

Abstract

This invention relates to the field of unmanned aerial vehicle (UAV) technology, specifically to a UAV servo flap system, a UAV wing, and the UAV itself. The system includes a flap actuator, an inertial measurement module, an airspeed detection module, and an onboard control module. The onboard control module fuses inertial data and airspeed data in real time through a data fusion unit, employing a composite filtering architecture combining a Kalman filter and a complementary filter. A dynamic flap angle calculation unit, based on an aerodynamic model and flight state parameters, uses a gradient descent method to optimize and solve for the optimal flap deflection angle. An error compensation unit corrects inertial measurement errors and the effects of structural elastic deformation online, thereby achieving accurate perception and real-time estimation of flight state, improving flap control accuracy and response speed, and enhancing the reliability of the UAV under complex flight conditions.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and more particularly to a UAV follow-up flap system, a UAV wing, and the UAV itself. Background Technology

[0002] As an important aerial platform, unmanned aerial vehicles (UAVs) are widely used in military reconnaissance, civilian surveying and mapping, agricultural plant protection, logistics and distribution, and environmental monitoring. With the continuous expansion of UAV application scenarios, higher requirements are being placed on their flight performance, maneuverability, economy, and safety. The UAV wing, as the main aerodynamic component generating lift, directly affects the UAV's flight performance. Flaps, as deflectable surfaces mounted on the trailing edge of the wing, are crucial devices for adjusting the UAV's aerodynamic characteristics. By changing the deflection angle of the flaps, the lift coefficient, drag coefficient, and pitch moment of the UAV can be effectively adjusted, thereby improving the performance of the UAV in different flight phases, including takeoff, cruise, maneuvering, and landing.

[0003] Traditional UAV flap systems primarily employ fixed-angle settings or simple open-loop control, where the flap deflection angle is pre-set based on the flight phase. This approach has significant limitations: First, fixed-angle flap settings cannot adapt to the complex and ever-changing flight conditions and aerodynamic environment during flight, resulting in a significant reduction in aerodynamic efficiency under off-design conditions. Second, simple open-loop control cannot perceive and respond to changes in the UAV's flight status in real time; the flap response lags behind these changes, limiting control accuracy and response speed. Third, traditional flap control does not fully consider the measurement error characteristics of the inertial measurement module and the impact of elastic deformation of the UAV structure under high-maneuverability flight conditions, leading to a deviation between the actual flap deflection angle and the commanded angle, reducing the reliability of the control system and flight safety. Summary of the Invention

[0004] The purpose of this invention is to provide a drone follow-up flap system, a drone wing, and a drone, which uses multi-source data to control the flap attitude to improve flight reliability.

[0005] To achieve the above objectives, the present invention provides a drone follow-up flap system, including a flap actuator disposed on the trailing edge of the drone wing, for driving the flap to rotate relative to the drone wing; An inertial measurement module is used to collect real-time three-axis angular velocity and acceleration data of the UAV; The airspeed detection module is used to collect the flight speed data of the drone in real time; The airborne control module is communicatively connected to the flap actuator, the inertial measurement module, and the airspeed detection module, respectively. The airborne control module includes a data fusion unit, a dynamic flap angle calculation unit, an error compensation unit, and an execution control unit. The data fusion unit is used to fuse the three-axis angular velocity, acceleration data and flight speed data in real time to generate a comprehensive feature vector of the current flight state of the UAV. The dynamic flap angle calculation unit is used to calculate the optimal flap deflection angle based on the comprehensive feature vector and the preset aerodynamic model. The error compensation unit is used to perform online error compensation for the optimal flap deflection angle based on the measurement error characteristics of the inertial measurement module and the elastic deformation parameters of the UAV structure. The execution control unit is used to send the flap deflection angle command after error compensation to the flap actuator.

[0006] The data fusion unit includes a data preprocessing subunit, a multi-source data alignment subunit, and an adaptive weighted fusion subunit. The data preprocessing subunit is connected to the inertial measurement module and the airspeed detection module respectively, and is used to perform noise filtering, outlier removal and time synchronization calibration on the triaxial angular velocity, acceleration data and flight speed data. The multi-source data alignment subunit is used to align the inertial data collected by the inertial measurement module with the aerodynamic data collected by the airspeed detection module by timestamp, and perform interpolation calculations based on a preset sampling rate to generate a unified data frame at the synchronization moment. The adaptive weighted fusion subunit includes a Kalman filter and a complementary filter: the Kalman filter is used to perform optimal estimation and fusion of inertial measurement data and flight velocity data based on the system state equation and observation equation to generate a fused state variable; the complementary filter is used to supplement and correct the fused state variable based on the three-axis angular velocity data when there is a delay or loss in the flight velocity data to obtain flight state data, and extract a multi-dimensional comprehensive feature vector based on the fused flight state data.

[0007] The Kalman filter includes: The state equation establishment sub-unit is used to establish the system state equation with UAV angular velocity, acceleration, velocity and position as state variables, as well as the observation equation with inertial measurement data and airspeed data as observation quantities; The prediction sub-unit is used to predict the prior state estimate and the corresponding error covariance matrix at the current moment based on the state estimate and system state equation at the previous moment. The update sub-unit is used to take the flight speed data collected in real time by the airspeed detection module as the observation input, calculate the Kalman gain, and combine the observation equation to correct the state prior estimate, and generate the fused state variables and their error covariance matrix. The multi-sensor fusion subunit is used to perform joint state estimation by combining the triaxial angular velocity and acceleration data from the inertial measurement module with the flight speed data, thereby eliminating the influence of single-sensor measurement noise and random errors.

[0008] The complementary filter includes: The high-frequency component extraction subunit is used to perform high-pass filtering on the triaxial angular velocity data output by the inertial measurement module to extract the high-frequency angular velocity components. The low-frequency component extraction subunit is used to perform low-pass filtering on the flight speed data to extract the low-frequency speed component. The delay detection subunit is used to monitor the update cycle of the flight speed data in real time. When the data delay exceeds a preset threshold or data is lost, the complementary correction mode is triggered. The fusion correction subunit is used to integrate the high-frequency angular velocity component to obtain the angle increment in the complementary correction mode, and then vector-superimpose it with the fusion state quantity output by the Kalman filter to generate the compensated fusion state quantity.

[0009] The dynamic flap angle calculation unit includes: The aerodynamic model sub-unit stores a preset aerodynamic model database, which is used to establish the mapping relationship between the flap deflection angle and the aerodynamic characteristic parameters of the UAV. A flight state parameter extraction subunit is used to extract the current flight state parameters from the comprehensive feature vector; The aerodynamic load calculation subunit is used to calculate the aerodynamic load distribution under the current flight conditions based on the extracted flight state parameters and the aerodynamic model database. The objective function construction sub-unit is used to construct an objective function with flap deflection angle as the optimization variable based on the aerodynamic load distribution according to the current flight mission requirements of the UAV. The objective function includes maximizing lift, minimizing drag, and balancing pitch moment. The constraint processing subunit is used to set the physical constraints for the flap deflection angle. An optimization sub-unit is used to solve the objective function under the physical constraints using the gradient descent method, and outputs the optimal flap deflection angle candidate value that satisfies all constraints.

[0010] The error compensation unit includes an inertial measurement error characteristic analysis subunit, a structural elastic deformation parameter estimation subunit, and an online error compensation execution subunit. The inertial measurement error characteristic analysis subunit is used to iteratively correct the inertial measurement error parameters by comparing the difference between the displacement increment calculated by inertial navigation and the position increment obtained by airspeed integration at the same time. The structural elastic deformation parameter estimation subunit is used to establish the mapping relationship between the elastic deformation amount and the installation position deviation of the flap actuator, and to calculate the deviation between the actual deflection angle of the flap and the command angle caused by the elastic deformation of the wing. The online error compensation execution subunit is used to add the optimal flap deflection angle to the deviation to obtain the error-compensated flap deflection angle command value, and output it to the execution control unit.

[0011] The specific steps of algebraically adding the optimal flap deflection angle to the error compensation value to obtain the error-compensated flap deflection angle command value, and then outputting it to the execution control unit, include: Based on the flight state parameters in the comprehensive feature vector, the aerodynamic load distribution acting on the UAV wing is estimated, including the bending moment, shear force and torque at the wing root; Based on the preset finite element model of the wing structure, the aerodynamic load is substituted into the calculation to obtain the elastic deformation of the wing, including the vertical displacement of the wingtip, the wingtip trailing edge twist angle, and the position deviation of the trailing edge point. Establish a mapping relationship between the elastic deformation amount and the installation position deviation of the flap actuator, and calculate the deviation between the actual flap deflection angle and the command angle caused by the elastic deformation of the wing.

[0012] The specific steps of adding the optimal flap deflection angle to the deviation to obtain the error-compensated flap deflection angle command value and outputting it to the execution control unit include: The total error of the optimal flap deflection angle is obtained by vector superposition of the inertial measurement error and the structural elastic deformation error. The error compensation value is calculated based on the total error, and the compensation direction is opposite to the total error direction. The compensation magnitude is proportional to the total error magnitude according to a preset coefficient. The proportional coefficient is dynamically adjusted according to the flight stage and maneuver type of the UAV, increasing the compensation weight during high-maneuver flight and decreasing the compensation weight during stable flight. The optimal flap deflection angle is algebraically added to the error compensation value to obtain the error-compensated flap deflection angle command value, which is then output to the execution control unit.

[0013] The present invention also provides a wing for a drone, including the aforementioned drone follow-up flap system.

[0014] The present invention also provides a drone, including the wings of the aforementioned drone.

[0015] This invention discloses a drone servo flap system, a drone wing, and the drone itself. By integrating an inertial measurement module, an airspeed detection module, and an onboard control module, this invention constructs a complete multi-sensor data acquisition and fusion processing system, achieving accurate perception and real-time estimation of the drone's flight status. The data fusion unit employs a composite filtering architecture combining a Kalman filter and a complementary filter. The Kalman filter utilizes the system state equation and observation equation for optimal estimation, while the complementary filter provides supplementary correction when airspeed data is delayed or lost, effectively improving the accuracy and reliability of flight status estimation.

[0016] This invention proposes an online error compensation mechanism for inertial measurement errors and structural elastic deformation. The inertial measurement error characteristic analysis subunit iteratively corrects error parameters by comparing inertial navigation calculations with airspeed integrals; the structural elastic deformation parameter estimation subunit calculates the elastic deformation caused by aerodynamic loads and its impact on flap deflection angle using a finite element model; and the online error compensation execution subunit performs vector superposition compensation on both types of errors and dynamically adjusts the compensation weights according to the flight phase. This multi-level, multi-source error compensation mechanism significantly improves the accuracy and reliability of flap control. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0018] Figure 1 This is a structural diagram of the first embodiment of the present invention.

[0019] Figure 2 This is a structural diagram of the airborne control module according to the first embodiment of the present invention.

[0020] Figure 3 This is a structural diagram of the data fusion unit in the first embodiment of the present invention.

[0021] Figure 4 This is a structural diagram of the dynamic flap angle calculation unit of the first embodiment of the present invention.

[0022] Figure 5 This is a structural diagram of the error compensation unit in the first embodiment of the present invention.

[0023] In the diagram: 1. Flap actuator; 2. Inertial measurement module; 3. Airspeed detection module; 4. Airborne control module; 41. Data fusion unit; 411. Data preprocessing subunit; 412. Multi-source data alignment subunit; 413. Adaptive weighted fusion subunit; 4131. Kalman filter; 4132. Complementary filter; 42. Dynamic flap angle calculation unit; 421. Aerodynamic model subunit; 422. Flight state parameter extraction subunit; 423. Aerodynamic load calculation subunit; 424. Objective function construction subunit; 425. Constraint processing subunit; 426. Optimization solution subunit; 43. Error compensation unit; 431. Inertial measurement error characteristic analysis subunit; 432. Structural elastic deformation parameter estimation subunit; 433. Online error compensation execution subunit; 44. Execution control unit. Detailed Implementation

[0024] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, but should not be construed as limiting the present invention.

[0025] First embodiment: Please see Figures 1-5 This embodiment provides a basic implementation of a UAV follow-up flap system. The system mainly consists of four parts: flap actuator 1, inertial measurement module 2, airspeed detection module 3, and airborne control module 4. These parts cooperate with each other to achieve intelligent follow-up control of the UAV flaps.

[0026] The flap actuator 1, located on the trailing edge of the UAV wing, is a mechanical actuator that performs flap deflection. The flap actuator 1 includes a drive motor, a reducer, a position sensor, and a mechanical transmission mechanism. The drive motor receives the flap deflection angle command from the onboard control module 4, reduces speed and increases torque through the reducer, and drives the mechanical transmission mechanism to rotate the flap around its axis. The position sensor detects the actual deflection angle of the flap in real time and sends a position feedback signal to the onboard control module 4, forming a closed-loop position control. The selection of the flap actuator 1 should consider the following technical specifications: the maximum output torque should meet the aerodynamic load requirements under high-maneuver flight conditions; the position control accuracy should be better than 0.1 degrees; the response bandwidth should meet the control requirements of high-speed maneuver flight; and the weight should be as light as possible to minimize the impact on the UAV's payload.

[0027] The inertial measurement module 2 is a key sensor for acquiring the inertial motion information of the UAV, including a three-axis gyroscope and a three-axis accelerometer. The inertial measurement module 2 is used to acquire the UAV's three-axis angular velocity and three-axis acceleration data in real time. The sampling frequency of the inertial measurement module 2 is typically set to 100Hz to 200Hz to meet the data update requirements of high-speed maneuvering flight. The inertial measurement module 2 should have low noise density and zero-bias stability to reduce the impact on the accuracy of flight state estimation. The raw data output by the inertial measurement module 2 contains systematic errors such as sensor inherent noise, zero-bias error, scaling factor error, and installation error. These errors need to be compensated and filtered by the data preprocessing subunit 411 before data fusion processing.

[0028] The airspeed detection module 3 is used to collect the UAV's flight speed data in real time, serving as a crucial data source for calculating aerodynamic loads and performing multi-sensor data fusion. The airspeed detection module 3 can be implemented using a pitot-tube airspeed detection module, a GPS receiver, or a combination of both. In high-precision applications, the airspeed detection module 3 can employ a multi-antenna GPS receiver to obtain accurate flight speed information through Doppler measurement. The update frequency of the airspeed detection module 3 is typically between 10Hz and 50Hz, which differs from the sampling frequency of the inertial measurement module 2, requiring time synchronization processing via the multi-source data alignment subunit 412.

[0029] The airborne control module 4 is the core control unit of the entire system. It uses a high-performance embedded processor, such as an ARM Cortex-A series processor or a DSP processor, as its core computing unit to meet the computing power requirements for real-time data fusion and optimization calculations. The airborne control module 4 communicates with the flap actuator 1, inertial measurement module 2, and airspeed detection module 3 via a CAN bus or RS48 bus. The airborne control module 4 integrates four functional modules: a data fusion unit 41, a dynamic flap angle calculation unit 42, an error compensation unit 43, and an execution control unit 44. These modules work collaboratively to achieve a complete closed loop from flight status perception to flap execution control.

[0030] The data fusion unit 41 is the core algorithm unit of the airborne control module 4, responsible for fusing inertial measurement data and airspeed data to generate a comprehensive feature vector of the UAV's flight state. (Refer to...) Figure 2 The data fusion unit 41 includes three sub-modules: a data preprocessing sub-unit 411, a multi-source data alignment sub-unit 412, and an adaptive weighted fusion sub-unit 413.

[0031] The data preprocessing subunit 411 is connected to both the inertial measurement module 2 and the airspeed detection module 3 to preprocess the acquired raw data. The specific steps of data preprocessing include: first, using Kalman filtering or low-pass filtering to denoise the raw data and remove high-frequency noise from the sensor; second, performing outlier detection on the denoised data, using the Grubbs criterion or 3σ criterion to identify and remove outlier data points; third, performing error compensation on the data according to preset sensor calibration parameters, including zero-bias compensation, scale factor compensation, and installation error compensation; and finally, performing time synchronization calibration on the inertial measurement data and airspeed data to establish a unified time reference.

[0032] The multi-source data alignment subunit 412 is responsible for aligning the inertial data collected by the inertial measurement module 2 with the aerodynamic data collected by the airspeed detection module 3 using timestamps. Since the sampling frequencies and update times of the inertial measurement module 2 and the airspeed detection module 3 are different, the multi-source data alignment subunit 412 performs interpolation calculations based on a preset sampling rate to generate a unified data frame at the synchronization time. In specific implementation, the multi-source data alignment subunit 412 uses the sampling time of the inertial measurement data as a reference and maps the airspeed data to the sampling time of the inertial measurement data using linear interpolation or spline interpolation methods to generate a time-synchronized data frame.

[0033] The adaptive weighted fusion subunit 413 is the core of the data fusion algorithm, comprising two filtering units: a Kalman filter 4131 and a complementary filter 4132. (Refer to...) Figure 3 The algorithm flow of the Kalman filter 4131 includes the following steps: First, the state equation establishment subunit establishes the system state equation with the UAV's angular velocity, acceleration, velocity, and position as state variables, and simultaneously establishes the observation equation with inertial measurement data and airspeed data as observation quantities; Second, the prediction subunit predicts the state prior estimate and the corresponding error covariance matrix at the current moment based on the state estimate value at the previous moment and the system state equation; Third, the update subunit uses the flight speed data collected in real time by the airspeed detection module 3 as the observation input, calculates the Kalman gain, and corrects the state prior estimate value in combination with the observation equation to generate the fused state quantity and its error covariance matrix; Fourth, the multi-sensor fusion subunit performs joint state estimation with the three-axis angular velocity and acceleration data of the inertial measurement module 2 and the flight speed data to eliminate the influence of single sensor measurement noise and random errors.

[0034] The complementary filter 4132 functions when there is delay or loss in flight velocity data. The working principle of the complementary filter 4132 is to utilize the characteristics of inertial measurement data (good high-frequency characteristics but with integral drift) and airspeed data (good low-frequency characteristics but with measurement delay), achieving a balance by weighted fusion of the high and low frequency components. The complementary filter 4132 includes a high-frequency component extraction subunit, a low-frequency component extraction subunit, a delay detection subunit, and a fusion correction subunit. The high-frequency component extraction subunit performs high-pass filtering on the three-axis angular velocity data output from the inertial measurement module 2 to extract the high-frequency angular velocity components. The low-frequency component extraction subunit performs low-pass filtering on the flight velocity data to extract the low-frequency velocity components. The delay detection subunit monitors the update cycle of the flight velocity data in real time. When a data delay exceeds a preset threshold or data loss is detected, the complementary correction mode is triggered. In the complementary correction mode, the fusion correction subunit integrates the high-frequency angular velocity components to obtain the angle increment, and then vector-superimposes it with the fused state quantity output by the Kalman filter 4131 to generate a compensated fused state quantity.

[0035] The comprehensive feature vector output by the data fusion unit 41 includes information such as the UAV's three-axis attitude angles, three-axis attitude angular velocities, three-axis positions, three-axis velocities, aerodynamic angle of attack, sideslip angle, and the estimation accuracy of each parameter. This information provides accurate flight status input for the dynamic flap angle calculation unit 42.

[0036] This embodiment provides a preferred implementation of the dynamic flap angle calculation unit 42.

[0037] The dynamic flap angle calculation unit 42 calculates the optimal flap deflection angle based on the comprehensive feature vector and the preset aerodynamic model, and is a key algorithm unit for realizing adaptive control of UAV flaps. The dynamic flap angle calculation unit 42 includes an aerodynamic model subunit 421, a flight state parameter extraction subunit 422, an aerodynamic load calculation subunit 423, an objective function construction subunit 424, a constraint condition processing subunit 425, and an optimization solution subunit 426.

[0038] The aerodynamic model subunit 421 stores a pre-set aerodynamic model database, which serves as the foundational data for aerodynamic characteristic calculations and flap angle optimization. This database is acquired through wind tunnel experiments, computational fluid dynamics simulations, or flight tests, establishing a mapping relationship between flap deflection angles and UAV aerodynamic characteristic parameters. The aerodynamic model database is typically stored using multidimensional interpolation tables or polynomial fitting, and includes the following aerodynamic characteristic data: curves showing the lift coefficient varying with flap deflection angles at different flight speeds; curves showing the drag coefficient varying with flap deflection angles at different flight speeds; and curves showing the pitch moment coefficient varying with flap deflection angles at different flight speeds. The aerodynamic characteristic data should cover various flight conditions that the UAV may encounter, including different flight speeds, altitudes, angles of attack, and sideslip angle ranges.

[0039] The flight state parameter extraction subunit 422 extracts the current flight state parameters from the comprehensive feature vector output by the data fusion unit 41. The extracted flight state parameters include: flight speed (airspeed or ground speed), flight altitude, angle of attack, sideslip angle, three-axis attitude angles, and three-axis attitude angular velocities. These flight state parameters will be used as inputs for aerodynamic load calculation.

[0040] The aerodynamic load calculation subunit 423 calculates the aerodynamic load distribution under the current flight conditions based on the extracted flight state parameters and the aerodynamic model database. Aerodynamic load calculation is a prerequisite for flap angle optimization, requiring accurate calculation of the aerodynamic forces and moments acting on the UAV wing. The specific steps of aerodynamic load calculation include: obtaining the lift coefficient, drag coefficient, and pitch moment coefficient from the aerodynamic model database based on the current flight speed, angle of attack, and flap deflection angle; calculating the atmospheric density based on the flight altitude; calculating the aerodynamic forces acting on the wing based on the UAV's wing area; integrating and distributing the aerodynamic forces along the wingspan to obtain the aerodynamic loads at various wing positions; and calculating the bending moment, shear force, and torque at the wing root to provide input for structural elastic deformation calculations.

[0041] Subunit 424, which constructs the objective function, builds an objective function with the flap deflection angle as the optimization variable based on the aerodynamic load distribution and the current flight mission requirements of the UAV. The construction of the objective function is the core of flap angle optimization and requires comprehensive consideration of multiple optimization objectives. The objective function includes the following three sub-objectives: maximizing lift, requiring the flap deflection angle to maximize lift for the UAV to meet the needs of takeoff, climb, and high-maneuverability flight; minimizing drag, requiring the flap deflection angle to minimize aerodynamic drag while meeting lift requirements, thereby improving the UAV's cruise efficiency and range; and balancing pitch moment, requiring the pitch moment generated by the flap deflection angle to balance the pitch moment generated by other UAV components to ensure the UAV's attitude stability.

[0042] Sub-unit 424 uses a weighted summation method to integrate multiple sub-objectives into a unified objective function, as shown below: Objective function: J(δ) = w1·JL(δ) - w2·JD(δ) + w3·JM(δ) Where δ is the flap deflection angle; JL(δ) is the lift-related term; JD(δ) is the drag-related term; JM(δ) is the pitch moment-related term; w1, w2, and w3 are the weight coefficients of each sub-target, and the weight coefficients are determined according to the flight mission requirements.

[0043] The constraint processing subunit 425 sets the physical constraints for the flap deflection angle, including upper and lower limits for the flap deflection angle, upper limit for the flap deflection angular velocity, and upper limit for the flap deflection acceleration. The physical constraints for the flap deflection angle depend on the structural design of the flap and the hinge moment characteristics. Typically, the upper limit for positive deflection angle is set to 2 to 40 degrees, and the lower limit for negative deflection angle is set to -10 to -5 degrees.

[0044] The optimization sub-unit 426 employs gradient descent to solve the objective function under physical constraints, outputting candidate values ​​for the optimal flap deflection angle that satisfy all constraints. The specific steps of gradient descent include: initializing the flap deflection angle; calculating the gradient of the objective function with respect to the flap deflection angle; updating the flap deflection angle according to the gradient direction; checking whether the updated flap deflection angle satisfies the constraints; if not, projecting it into the feasible region; repeating the above steps until the convergence condition is met. The convergence threshold and maximum number of iterations for the optimization solution can be configured based on onboard computing resources, achieving real-time computation while ensuring solution accuracy.

[0045] The error compensation unit 43 is one of the core innovations of this invention. It is used to perform online error compensation on the optimal flap deflection angle output by the dynamic flap angle calculation unit 42 based on the measurement error characteristics of the inertial measurement module 2 and the elastic deformation parameters of the UAV structure. The error compensation unit 43 includes an inertial measurement error characteristic analysis subunit 431, a structural elastic deformation parameter estimation subunit 432, and an online error compensation execution subunit 433.

[0046] The inertial measurement error characteristic analysis subunit 431 is used to iteratively correct inertial measurement error parameters through comparative analysis. The measurement errors of the inertial measurement module 2 mainly include zero-bias error, scaling factor error, noise error, and random walk error. These errors drift with changes in flight time and environmental conditions, requiring online estimation and compensation. The specific working principle of the inertial measurement error characteristic analysis subunit 431 is as follows: By comparing the difference between the displacement increment calculated by inertial navigation and the position increment obtained by airspeed integration at the same moment, the cumulative information of inertial measurement error can be obtained. Since the position increment obtained by airspeed integration does not include the influence of inertial measurement error, the difference between the two position increments is the integral result of the inertial measurement error. By inversely solving this error integration result, the inertial measurement error parameters, including the zero-bias estimate and scaling factor correction coefficient, can be iteratively corrected. The inertial measurement error characteristic analysis subunit 431 starts working after the UAV takes off and enters the stable flight phase, updating the error parameters every preset time interval.

[0047] The structural elastic deformation parameter estimation subunit 432 is used to calculate the deviation between the actual flap deflection angle and the commanded angle caused by the elastic deformation of the wing. Under high-maneuver flight conditions, the UAV's wing experiences significant aerodynamic loads and undergoes elastic deformation. This elastic deformation affects the installation position and attitude of the flap actuator 1, leading to a deviation between the actual flap deflection angle and the commanded angle. The specific working steps of the structural elastic deformation parameter estimation subunit 432 are as follows.

[0048] The first step is to establish a finite element model of the wing structure. This finite element model discretizes the wing structure into several nodes and elements. The nodes include the installation nodes of the flap actuator 1, characteristic locations such as the wing root and wingtip, and the element properties include bending stiffness and torsional stiffness.

[0049] The second step is to estimate the aerodynamic load distribution acting on the UAV wing based on the flight state parameters in the comprehensive feature vector. The wing root bending moment, shear force, and torque output by the aerodynamic load calculation subunit 423 are used as the external load inputs to the finite element model.

[0050] The third step involves substituting the aerodynamic loads into the static equations based on the finite element model of the wing structure to obtain the elastic deformation of the wing, including the vertical displacement of the wingtip, the torsion angle of the wingtip trailing edge, and the position deviation of the trailing edge point.

[0051] The fourth step is to establish a mapping relationship between the amount of elastic deformation and the installation position deviation of the flap actuator 1. This mapping relationship can be obtained by fitting finite element simulation data or by approximate calculation using analytical formulas. Based on this mapping relationship, the deviation between the actual flap deflection angle and the commanded angle caused by the elastic deformation of the wing is calculated.

[0052] The online error compensation execution subunit 433 adds the optimal flap deflection angle to the deviation to obtain the error-compensated flap deflection angle command value, and outputs it to the execution control unit 44. The specific workflow of the online error compensation execution subunit 433 includes the following steps: The total error of the optimal flap deflection angle is obtained by vector superposition of inertial measurement error and structural elastic deformation error. Inertial measurement error mainly affects the accuracy of flight state estimation, and thus the accuracy of aerodynamic load calculation; structural elastic deformation error directly affects the physical deflection angle of the flap. Since the two errors have different mechanisms of action, the total error needs to be calculated by superposition in vector space.

[0053] The error compensation value is calculated based on the total error. The calculation of the error compensation value follows these principles: the compensation direction is opposite to the direction of the total error to achieve error cancellation; the compensation magnitude is proportional to the magnitude of the total error by a preset coefficient, which is tuned according to the error characteristics and control requirements.

[0054] The proportional coefficient is dynamically adjusted based on the UAV's flight stage and maneuver type. Under high-maneuver flight conditions, the wings bear significant aerodynamic loads, resulting in a substantial increase in structural elastic deformation; in this case, the compensation weight for structural elastic deformation error is increased. Under stable flight conditions, the impact of inertial measurement error on control accuracy is relatively prominent; in this case, the compensation weight for inertial measurement error is increased. The flight stage and maneuver type can be determined from parameters such as angular velocity and acceleration extracted from the comprehensive feature vector.

[0055] The optimal flap deflection angle is algebraically added to the error compensation value to obtain the error-compensated flap deflection angle command value. This command value comprehensively considers both flight state optimization and error compensation, enabling precise flap control.

[0056] The execution control unit 44 receives the flap deflection angle command value after error compensation and sends it to the flap actuator 1 through the communication interface. The execution control unit 44 also receives the position feedback signal from the flap actuator 1 to form a closed-loop position control, ensuring that the actual flap deflection angle tracks the command value.

[0057] The control process of the UAV follow-up flap system includes the following steps.

[0058] Step S1: After the UAV is powered on, the onboard control module 4 performs self-test and initialization, including initialization of each sensor module, establishment of communication links, loading of the aerodynamic model database, and configuration of control parameters. After initialization is complete, the system enters normal operation.

[0059] In step S2, the inertial measurement module 2 collects the three-axis angular velocity and three-axis acceleration data of the UAV in real time, with a sampling frequency set to 100Hz. The airspeed detection module 3 collects the flight speed data of the UAV in real time, with an update frequency set to 20Hz. During the data acquisition process, the timestamps of the data from each sensor are recorded synchronously.

[0060] In step S3, the data preprocessing subunit 411 of the data fusion unit 41 performs denoising filtering, outlier removal, and time synchronization calibration on the acquired raw data. The denoising filtering uses a Kalman filter or low-pass filter algorithm, with the cutoff frequency set to 0.5 times the sensor bandwidth. Outlier detection uses the Grubbs criterion, with a confidence level set to 9%.

[0061] In step S4, the multi-source data alignment subunit 412 of the data fusion unit 41 timestamps the inertial measurement data and the airspeed data, and performs linear interpolation on the airspeed data based on the sampling time of the inertial measurement data to generate a unified data frame with time synchronization.

[0062] In step S5, the adaptive weighted fusion subunit 413 of the data fusion unit 41 fuses the time-synchronized inertial measurement data and airspeed data. The Kalman filter 4131 uses the system state equation and observation equation to perform optimal estimation fusion; the complementary filter 4132 provides supplementary correction when airspeed data is delayed or lost. The fusion output is a comprehensive feature vector of the UAV's current flight state.

[0063] Step S6: The flight state parameter extraction subunit 422 of the dynamic flap angle calculation unit 42 extracts the current flight state parameters from the comprehensive feature vector, including flight speed, flight altitude, flight angle of attack, sideslip angle, three-axis attitude angle and three-axis attitude angular velocity, etc.

[0064] Step S7: The aerodynamic load calculation subunit 423 of the dynamic flap angle calculation unit 42 calculates the aerodynamic load distribution under the current flight conditions based on the flight state parameters and aerodynamic model database, including lift, drag, pitching moment, and bending moment, shear force and torque at the wing root.

[0065] Step S8: The objective function construction subunit 424 of the dynamic flap angle calculation unit 42 constructs the objective function according to the flight mission requirements, the constraint condition processing subunit 425 sets the physical constraints, and the optimization solution subunit 426 uses the gradient descent method to solve for the optimal flap deflection angle candidate value.

[0066] Step S9: The inertial measurement error characteristic analysis subunit 431 of the error compensation unit 43 compares the displacement increment calculated by the inertial navigation with the airspeed integral displacement increment, and iteratively corrects the inertial measurement error parameters, including the zero bias estimate and the scaling factor correction coefficient.

[0067] In step S10, the structural elastic deformation parameter estimation subunit 432 of the error compensation unit 43 establishes a mapping relationship between the elastic deformation amount and the flap deflection angle deviation based on the elastic deformation amount of the wing under aerodynamic load, and calculates the flap deflection angle deviation caused by the elastic deformation of the wing.

[0068] In step S11, the online error compensation execution subunit 433 of the error compensation unit 43 superimposes the inertial measurement error and the structural elastic deformation error to obtain the total error, calculates the error compensation value, and dynamically adjusts the compensation weight according to the flight phase.

[0069] In step S12, the execution control unit 44 sends the error-compensated flap deflection angle command to the flap actuator 1, and the flap actuator 1 drives the flaps to deflect to the target angle. Simultaneously, the execution control unit 44 receives flap position feedback, forming a closed-loop control.

[0070] In step S13, the system determines whether the drone is still running. If it continues to run, it returns to step S2 to start the next control cycle. If the drone stops running, the system exits the control process.

[0071] The above control process is executed cyclically with a fixed sampling period and a sampling control frequency of 20Hz. Within each control cycle, the system completes the entire process from data acquisition to control execution, achieving real-time follow-up control of the UAV flaps.

[0072] This invention also provides a wing for a drone, including a drone follow-up flap system. The drone wing is the main aerodynamic component generating lift, and the follow-up flap system described in this invention is integrated into the existing structure. The structural design of the drone wing needs to consider the installation arrangement of the flap actuator 1, the inertial measurement module 2, the airspeed detection module 3, and the onboard control module 4. The flap actuator 1 is installed in the flap mounting area on the trailing edge of the wing and is connected to the flap via a mechanical transmission mechanism; the inertial measurement module 2 is installed inside the wing near the center of gravity to obtain accurate attitude measurement data; the airspeed detection module 3 is installed in the undisturbed airflow area on the leading edge of the wing or the nose of the fuselage; the onboard control module 4 is installed in the electronics bay inside the wing and is connected to the various sensors and actuators via cables. The drone wing also has necessary electrical and mechanical interfaces for electrical connection and mechanical fixation to the drone fuselage.

[0073] This invention also provides an unmanned aerial vehicle (UAV), including the wings of the aforementioned UAV. The UAV wings are symmetrically mounted on both sides of the UAV fuselage, forming the UAV's lifting surface system. The power system provides thrust to the UAV and can be in the form of an electric propeller or an internal combustion engine driving the propeller. The flight control system is responsible for the UAV's attitude stabilization and trajectory control, working in conjunction with the servo flap system of this invention to achieve overall optimization of the UAV's flight performance. The payload is configured according to mission requirements and may include aerial survey cameras, infrared thermal imagers, communication equipment, transported goods, etc.

[0074] The above-disclosed embodiments are merely one or more preferred embodiments of this application and should not be construed as limiting the scope of this application. Those skilled in the art can understand that all or part of the processes for implementing the above embodiments and equivalent changes made in accordance with the claims of this application still fall within the scope of this application.

Claims

1. A drone follow-up flap system, characterized in that, Includes a flap actuator, located at the trailing edge of the UAV wing, used to drive the flaps to rotate relative to the UAV wing; An inertial measurement module is used to collect real-time three-axis angular velocity and acceleration data of the UAV; The airspeed detection module is used to collect the flight speed data of the drone in real time; The airborne control module is communicatively connected to the flap actuator, the inertial measurement module, and the airspeed detection module, respectively. The airborne control module includes a data fusion unit, a dynamic flap angle calculation unit, an error compensation unit, and an execution control unit. The data fusion unit is used to fuse the three-axis angular velocity, acceleration data and flight speed data in real time to generate a comprehensive feature vector of the current flight state of the UAV. The dynamic flap angle calculation unit is used to calculate the optimal flap deflection angle based on the comprehensive feature vector and the preset aerodynamic model. The error compensation unit is used to perform online error compensation for the optimal flap deflection angle based on the measurement error characteristics of the inertial measurement module and the elastic deformation parameters of the UAV structure. The execution control unit is used to send the flap deflection angle command after error compensation to the flap actuator.

2. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 1, characterized in that, The data fusion unit includes a data preprocessing subunit, a multi-source data alignment subunit, and an adaptive weighted fusion subunit; The data preprocessing subunit is connected to the inertial measurement module and the airspeed detection module respectively, and is used to perform noise reduction filtering, outlier removal and time synchronization calibration on the triaxial angular velocity, acceleration data and flight speed data. The multi-source data alignment subunit is used to align the inertial data collected by the inertial measurement module with the aerodynamic data collected by the airspeed detection module by timestamp, and perform interpolation calculations based on a preset sampling rate to generate a unified data frame at the synchronization moment. The adaptive weighted fusion subunit includes a Kalman filter and a complementary filter: the Kalman filter is used to perform optimal estimation and fusion of inertial measurement data and flight velocity data based on the system state equation and observation equation to generate a fused state variable; the complementary filter is used to supplement and correct the fused state variable based on the three-axis angular velocity data when there is a delay or loss in the flight velocity data to obtain flight state data, and extract a multi-dimensional comprehensive feature vector based on the fused flight state data.

3. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 2, characterized in that, The Kalman filter includes: The state equation establishment sub-unit is used to establish the system state equation with UAV angular velocity, acceleration, velocity and position as state variables, as well as the observation equation with inertial measurement data and airspeed data as observation quantities; The prediction sub-unit is used to predict the prior state estimate and the corresponding error covariance matrix at the current moment based on the state estimate and system state equation at the previous moment. The update sub-unit is used to take the flight speed data collected in real time by the airspeed detection module as the observation input, calculate the Kalman gain, and correct the state prior estimate by combining the observation equation to generate the fused state variables and their error covariance matrix. The multi-sensor fusion subunit is used to perform joint state estimation by combining the triaxial angular velocity and acceleration data from the inertial measurement module with the flight speed data, thereby eliminating the influence of single-sensor measurement noise and random errors.

4. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 3, characterized in that, The complementary filter includes: The high-frequency component extraction subunit is used to perform high-pass filtering on the triaxial angular velocity data output by the inertial measurement module to extract the high-frequency angular velocity components. The low-frequency component extraction subunit is used to perform low-pass filtering on the flight speed data to extract the low-frequency speed component. The delay detection subunit is used to monitor the update cycle of the flight speed data in real time. When the data delay exceeds a preset threshold or data is lost, the complementary correction mode is triggered. The fusion correction subunit is used to integrate the high-frequency angular velocity component to obtain the angle increment in the complementary correction mode, and then vector-superimpose it with the fusion state quantity output by the Kalman filter to generate the compensated fusion state quantity.

5. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 4, characterized in that, The dynamic flap angle calculation unit includes: The aerodynamic model sub-unit stores a preset aerodynamic model database, which is used to establish the mapping relationship between the flap deflection angle and the aerodynamic characteristic parameters of the UAV. A flight state parameter extraction subunit is used to extract the current flight state parameters from the comprehensive feature vector; The aerodynamic load calculation subunit is used to calculate the aerodynamic load distribution under the current flight conditions based on the extracted flight state parameters and the aerodynamic model database. The objective function construction sub-unit is used to construct an objective function with flap deflection angle as the optimization variable based on the aerodynamic load distribution according to the current flight mission requirements of the UAV. The objective function includes maximizing lift, minimizing drag, and balancing pitch moment. The constraint processing subunit is used to set the physical constraints for the flap deflection angle. An optimization sub-unit is used to solve the objective function under the physical constraints using the gradient descent method, and outputs the optimal flap deflection angle candidate value that satisfies all constraints.

6. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 5, characterized in that, The error compensation unit includes an inertial measurement error characteristic analysis subunit, a structural elastic deformation parameter estimation subunit, and an online error compensation execution subunit. The inertial measurement error characteristic analysis subunit is used to iteratively correct the inertial measurement error parameters by comparing the difference between the displacement increment calculated by inertial navigation and the position increment obtained by airspeed integration at the same time. The structural elastic deformation parameter estimation subunit is used to establish the mapping relationship between the elastic deformation amount and the installation position deviation of the flap actuator, and to calculate the deviation between the actual deflection angle of the flap and the command angle caused by the elastic deformation of the wing. The online error compensation execution subunit is used to add the optimal flap deflection angle to the deviation to obtain the error-compensated flap deflection angle command value, and output it to the execution control unit.

7. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 6, characterized in that, The specific steps of algebraically adding the optimal flap deflection angle to the error compensation value to obtain the error-compensated flap deflection angle command value, and outputting it to the execution control unit, include: Based on the flight state parameters in the comprehensive feature vector, the aerodynamic load distribution acting on the UAV wing is estimated, including the bending moment, shear force and torque at the wing root; Based on the preset finite element model of the wing structure, the aerodynamic load is substituted into the calculation to obtain the elastic deformation of the wing, including the vertical displacement of the wingtip, the wingtip trailing edge twist angle, and the position deviation of the trailing edge point. Establish a mapping relationship between the elastic deformation amount and the installation position deviation of the flap actuator, and calculate the deviation between the actual flap deflection angle and the command angle caused by the elastic deformation of the wing.

8. The unmanned aerial vehicle (UAV) follow-up flap system as described in claim 7, characterized in that, The specific steps of adding the optimal flap deflection angle to the deviation to obtain the error-compensated flap deflection angle command value and outputting it to the execution control unit include: The total error of the optimal flap deflection angle is obtained by vector superposition of the inertial measurement error and the structural elastic deformation error. The error compensation value is calculated based on the total error, and the compensation direction is opposite to the total error direction. The compensation magnitude is proportional to the total error magnitude according to a preset coefficient. The proportional coefficient is dynamically adjusted according to the flight stage and maneuver type of the UAV, increasing the compensation weight during high-maneuver flight and decreasing the compensation weight during stable flight. The optimal flap deflection angle is algebraically added to the error compensation value to obtain the error-compensated flap deflection angle command value, which is then output to the execution control unit.

9. A wing of a drone, characterized in that, Includes a drone follow-up flap system as described in any one of claims 1 to 8.

10. A drone, characterized in that, Includes the wing of a drone as described in claim 9.

Images