Method and system for dynamically controlling the speed of a heavy unmanned aerial vehicle according to the wind

Abstract

The application belongs to the field of flight control technology, and particularly relates to a method and system for dynamically controlling the rotating speed of a heavy unmanned aerial vehicle (UAV) according to wind force, a sensor module, a data processing and solving module, an adaptive anti-disturbance controller, and an actuating mechanism. The sensor module is used for acquiring the state information of the UAV in real time, and the state information includes attitude information, position information and speed information under the shaft system of the UAV body. The data processing and solving module is used for collecting the data of the sensor module and running a flight control algorithm. The adaptive anti-disturbance controller is used for receiving a desired instruction and the state information, and outputting a control quantity. The actuating mechanism is used for receiving the control quantity, driving the rotors and control surfaces of the UAV, and inhibiting the influence of external disturbance on the flight state. The application is aimed at the strong coupling characteristics of the longitudinal double-rotor layout, a dynamic decoupling unit based on the invariance principle is designed, the mutual interference between the longitudinal and lateral channels is significantly weakened in structure, and the difficulty of subsequent controller design is simplified.

Description

Technical Field

[0001] This invention belongs to the field of flight control technology, specifically referring to a method and system for dynamically controlling the rotational speed of a heavy unmanned aerial vehicle (UAV) based on wind power. Background Technology

[0002] Tandem-rotor unmanned helicopters are widely used in military and civilian fields due to their excellent transport capacity and hovering efficiency. However, this configuration suffers from severe aerodynamic interference between the front and rear rotors, strong dynamic coupling between the longitudinal and lateral channels, and high system nonlinearity. Furthermore, in actual flight, they also face external disturbances such as measurement noise, trim errors, aerodynamic parameter perturbations, and gusts, making the design of a high-performance flight control system extremely challenging.

[0003] Traditional control methods, such as classical PID control, while simple in structure, are cumbersome in parameter tuning, have limited control performance for strongly coupled nonlinear systems, and exhibit poor robustness. Some modern control methods theoretically offer better performance, but their design is complex, highly dependent on model accuracy, and presents difficulties in practical engineering implementation and parameter tuning. Model reference adaptive control (MRAC) provides an effective way to address system uncertainties, but the parameter convergence and tracking accuracy of traditional MRAC structures are affected by significant measurement noise. Summary of the Invention

[0004] In view of the above situation and to overcome the shortcomings of the prior art, the present invention provides a method and system for dynamically controlling the rotation speed of a heavy unmanned aerial vehicle based on wind power.

[0005] The technical solution adopted by the present invention is as follows: The present invention provides a system for dynamically controlling the rotation speed of a heavy unmanned aerial vehicle (UAV) based on wind power, and a sensor module for acquiring the UAV's status information in real time, including attitude information, position information and speed information under the body axis; The data processing and calculation module is used to collect data from the sensor module and run flight control algorithms. An adaptive disturbance rejection controller is used to receive desired commands and status information and output control quantities. The actuator is used to receive control signals, drive the rotor and control surfaces of the UAV, and suppress the impact of external disturbances on the flight status.

[0006] Furthermore, the data processing and calculation module is a dual-core controller based on DSP and ARM. The ARM core is responsible for collecting data from the sensor module, while the DSP core is used to run the flight control algorithm.

[0007] Furthermore, the adaptive disturbance rejection controller adopts a structure that combines a reference model and a follower model, and also includes a variable gain module for improving parameter convergence.

[0008] Furthermore, the sensor module includes: The IMU (In-Mechanical Unit) attitude system is used to measure the attitude angles, angular rates, and linear acceleration of the UAV. Differential GPS is used to obtain high-precision position and ground speed information for drones; A barometer is used to provide altitude information and integrate it with differential GPS information.

[0009] Furthermore, the adaptive disturbance rejection controller includes: The follower model is used to take posture commands as input and output the desired posture response. Reference model, used as input for processed command signals; An error compensator is used to connect the following model and the controlled object; The variable gain module is used to dynamically adjust the gain of the error compensator according to the integral of the attitude tracking error, so as to achieve fast response in the early stage of control and stable convergence in the later stage of control.

[0010] Furthermore, the variable gain module is determined by the following formula: ; in, For attitude tracking error, and These are preset parameters.

[0011] Furthermore, the adaptive disturbance rejection controller also includes a dynamic decoupling unit, which constructs a decoupling network based on the principle of invariance, enabling independent control of the pitch and roll channels.

[0012] Furthermore, the implementing agencies include: Brushless motors and electronic speed controllers are used to drive rotors to maintain a constant or controlled speed; Multiple digital servos are used to manipulate the forward and backward automatic swashplates, which work together to change the collective pitch and cyclic pitch of the rotor to achieve attitude and trajectory control.

[0013] Furthermore, it also includes a position outer loop controller and a ground remote control and telemetry station. The position outer loop controller operates as the outer loop within the DSP core, receiving position, speed, or altitude commands, and generating inner loop attitude commands based on the deviation from the feedback information from the sensor modules. These commands are then transmitted to the adaptive disturbance rejection controller, forming a cascaded control structure. The ground remote control and telemetry station communicates with the data processing and calculation module via a wireless data transmission radio. It is used to send flight mode commands, waypoint tasks, and control parameters, receive and display UAV status data and flight trajectory in real time, record flight data, and send remote takeover commands in emergency situations.

[0014] This solution also discloses the operation method of a wind-driven dynamic control system for the rotational speed of heavy-duty unmanned aerial vehicles, which mainly includes the following steps: Step A1: Obtain real-time status information of the drone through the sensor module; Step A2: In the DSP core of the dual-core controller, run the adaptive disturbance rejection control law. The control law adopts a structure that combines the reference model and the follower model, and uses a variable gain strategy to adjust the error compensation. Step A3: The control law calculates the control quantity for the actuator based on the deviation between the expected command and the state information; Step A4: Drive the drone through the actuator to stably track the desired flight state in an environment with uncertainties such as wind disturbance.

[0015] The beneficial effects achieved by the present invention using the above structure are as follows: The present invention provides a system for dynamically controlling the rotational speed of a heavy-duty unmanned aerial vehicle based on wind power, achieving the following beneficial effects: (1) In view of the strong coupling characteristics of the tandem dual rotor layout, a dynamic decoupling unit based on the invariance principle was designed, which significantly reduced the mutual interference between the longitudinal and lateral channels from the structure, and simplified the design difficulty of the subsequent controller. In the specific implementation process, the nonlinear dynamic model of the UAV near the hovering balance point was linearized first to establish the state space model of the system, and the transfer function matrix between the longitudinal and lateral control inputs was obtained by Laplace transform. In this transfer function matrix, the off-diagonal elements represent the coupling relationship between different control channels of the system. In order to eliminate this coupling effect, a dynamic decoupling compensation network was introduced into the control system. The original control input was transformed by designing a decoupling compensator, so that the original multi-input multi-output system with coupling relationship was approximately transformed into multiple independent single-input single-output systems. In this way, independent control between the pitch channel and the roll channel can be realized, thereby significantly reducing the mutual interference between control channels. The dynamic decoupling algorithm is run in real time in the DSP processor in the form of a digital filter in the actual system and is located before the servo control command output to process the control signal in real time. Through the above structural design, the coupling effect between the longitudinal and lateral channels can be effectively reduced, thereby simplifying the design difficulty of the subsequent attitude controller.

[0016] (2) An improved model reference adaptive control framework is adopted, which can estimate and compensate for system uncertainties caused by changes in aerodynamic parameters and inaccurate trimming online, so that the control system has inherent robustness to model errors and internal disturbances. In specific implementation, a reference model is first constructed to describe the dynamic response characteristics of the UAV under ideal conditions. The reference model usually adopts the form of a standard second-order system. Its input is the attitude control command and its output is the desired attitude response. The control system compares the actual attitude measured by the UAV with the desired attitude output by the reference model to obtain the attitude tracking error. Based on the error, a feedback control law is constructed. The control input is continuously adjusted through the error compensator so that the actual attitude of the UAV gradually approaches the ideal response described by the reference model. At the same time, in order to cope with the possible changes in the dynamic parameters of the UAV during flight, the system constructs a regression vector related to the system state and adjusts the controller parameters in real time through the adaptive parameter update law, so that the controller can estimate the dynamic characteristics of the system online and compensate for model errors. Through this adaptive adjustment mechanism, the control system can effectively cope with uncertainties such as changes in aerodynamic parameters, load changes and wind disturbances, thereby improving the robustness of the system to model errors and internal disturbances.

[0017] (3) A variable gain module is introduced, which cleverly combines the rapid response advantage of the error compensator in the early stage of control with the stable convergence advantage of the parameters in the later stage of control through its "fade-in and fade-out" characteristics. This overcomes the problem of parameter chattering in traditional adaptive control under noise and achieves fast, accurate and stable attitude tracking. By performing integral calculation on the attitude tracking error within a certain time window, the controller gain is dynamically adjusted according to the error integral value. When the system is in the early stage of control or the attitude error is large, the variable gain module will increase the controller gain, thereby enhancing the control effect and enabling the system to respond quickly and reduce the attitude error rapidly. When the system gradually approaches the steady state and the error decreases, the controller gain will gradually decrease to reduce the influence of noise on the parameter update process, thereby suppressing the oscillation phenomenon generated by the adaptive parameters in the steady state stage. Through this gain adjustment mechanism with "fade-in and fade-out" characteristics, a fast response speed can be maintained in the early stage of control, while improving the stability of parameter convergence when the system approaches the steady state. This method effectively overcomes the problem of parameter chattering in traditional adaptive control under measurement noise, enabling the system to achieve fast, accurate and stable attitude tracking.

[0018] (4) The DSP+ARM dual-core controller architecture is adopted, which separates data acquisition from complex control algorithm calculation. It makes full use of the advantages of the rich ARM interface and the strong numerical calculation capability of DSP, ensuring the high reliability and high real-time performance of the entire flight control system, and providing a hardware foundation for the engineering implementation of complex algorithms. The ARM processor is mainly responsible for sensor data acquisition and communication management. It reads the data output by IMU and differential GPS through the serial port interface, reads the barometer altitude information through the I²C bus, and receives the remote control control signal through the timer capture channel. After completing the data acquisition and preprocessing, the ARM processor writes the sorted data into the dual-port RAM to realize data sharing. The DSP processor reads the system status data from the dual-port RAM through the SPI interface. The system executes flight control algorithms, including dynamic decoupling algorithms, adaptive control laws, and variable gain adjustment strategies, to calculate control commands. These commands are then output as servo control signals and motor speed control signals via a PWM module. Dual-port RAM serves as a shared data storage area between the ARM and DSP, allowing both processors to access the same data area simultaneously without bus conflicts, thus ensuring real-time data exchange. This dual-core architecture, where the DSP and ARM work together, fully leverages the advantages of the ARM in interface management and data acquisition, as well as the DSP's high-speed numerical computation capabilities, significantly improving the real-time performance and reliability of the flight control system. This provides a reliable hardware foundation for the engineering implementation of complex control algorithms in UAV systems. Attached Figure Description

[0019] Figure 1 The flowchart of the wind-power dynamic control system for heavy-duty unmanned aerial vehicle (UAV) speed proposed in this invention is shown below. Figure 2 This is the internal flowchart of the adaptive disturbance rejection controller; Figure 3 Flowchart for variable gain adaptive adjustment; Figure 4 This is a flowchart of the signal processing for the dynamic decoupling unit. Figure 5 This is a flowchart of the DSP + ARM dual-core collaborative workflow.

[0020] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: Please see Figures 1-5 As shown, this embodiment is a wind-power dynamic control system for the rotational speed of a heavy-duty UAV, including a sensor module, a data processing and calculation module, an adaptive anti-disturbance controller, and an actuator.

[0023] The sensor module is used to acquire the status information of the UAV in real time. The status information includes attitude angle, angular rate and linear acceleration measured by IMU, position and altitude information obtained by fusion of differential GPS and barometer, and velocity information under the body axis. The data processing and calculation module is a dual-core controller based on DSP and ARM. The ARM core is responsible for collecting data from the sensor module, and the DSP core is used to run flight control algorithms. An adaptive disturbance rejection controller, which operates as the core control law in the DSP core, adopts a structure combining a reference model and a follower model, and includes a variable gain module. The controller receives the desired command and the state information, dynamically adjusts the control action through the variable gain module, and outputs the control quantity. The actuator receives the control input and drives the rotor and control surfaces of the UAV to suppress the influence of external disturbances on the flight status.

[0024] The adaptive disturbance rejection controller also includes a dynamic decoupling unit, which is designed based on the invariance principle for the longitudinal and lateral coupling characteristics of the tandem dual rotor configuration. The dynamic decoupling unit controls the longitudinal and lateral channels independently through the decoupling network.

[0025] Sensor selection and integration: The IMU used is MTi-300, which outputs 100Hz attitude, angular rate and acceleration data via RS-232 serial port.

[0026] The DGPS uses the OEM617D board and an antenna to output 10Hz RTK positioning and speed data via an RS-232 serial port.

[0027] The barometer is an MS5611, which communicates with the main controller via the I²C bus.

[0028] Dual-core flight control computer design: The core uses a TMS320F28335 DSP and an STM32F407 ARM Cortex-M4 MCU.

[0029] The ARM minimum system is responsible for: acquiring IMU and DGPS data through multiple UART interfaces; reading barometer data through the I²C interface; receiving remote control PPM signals through the timer capture channel; and communicating with the dual-port RAM through the SPI interface.

[0030] The DSP minimum system is responsible for: reading all processed sensor data from the dual-port RAM via the SPI interface; running the adaptive disturbance rejection control algorithm; and generating 6 servo control signals and 1 motor speed control signal through its enhanced PWM module.

[0031] The dual-port RAM uses the IDT71321 as a global data area shared by the ARM and DSP, enabling conflict-free data exchange between cores.

[0032] Implementing agency: The brushless motor is an Acron 700MX, paired with a Hobbywing 200A high-voltage electronic speed controller.

[0033] The digital servo uses six Altec Lansing DS610s to control the three control points of the forward and backward automatic tilting devices.

[0034] Communication link: The ground station and the airborne system communicate using Digi's Xtend 900MHz data radio.

[0035] 1. Algorithm formula for dynamic decoupling unit: 1) Linearized model and transfer function matrix: Near the hovering equilibrium point, the nonlinear dynamics model of the helicopter is linearized to obtain the state-space model: ; in, ; Define the longitudinal and lateral control inputs as follows: , ; The longitudinal and transverse transfer function matrices are obtained through Laplace transform: ; Function: Used to describe the dynamic behavior of UAVs near the hovering equilibrium point; by linearizing the nonlinear model, a state-space model is obtained, and then the transfer function matrix between the longitudinal and lateral channels is derived, providing a mathematical model basis for subsequent decoupled control.

[0036] 2) Decoupling compensator design formula To eliminate longitudinal and lateral coupling, a dynamic decoupling network is introduced: ; The decoupled system is required to satisfy: ; This leads to the ideal decoupling compensator: ; Purpose: To design a dynamic decoupling network so that the transfer function matrix between the system input and output is a diagonal matrix, thereby enabling independent control of the longitudinal and transverse channels.

[0037] 3) Low-order approximation and digital implementation Will , Perform low-order approximations: , ; Discretization using bilinear transform: ; The digital filter form is obtained as follows: ; This decoupling network operates in real time at the front end of the servo command output in the DSP.

[0038] Function: To approximate the continuous-time decoupling compensator as a low-order transfer function, making it easier to implement in DSP; to discretize it through bilinear transformation, obtaining a recursive formula in the form of a digital filter for real-time control.

[0039] 2. Improved Adaptive Attitude Controller 1) Follower Model Selecting a second-order reference model: ; Parameter values: , , ; Corresponding time-domain model: ; Function: Define the desired attitude response characteristics as the target model for the controller to track.

[0040] 2) Attitude error and filtering error Define attitude tracking error: ; Filtering error: ; Function: Calculate the error between the actual attitude and the desired attitude; introduce filtering error for adaptive law design to enhance the system's robustness to noise.

[0041] 3) Variable gain adjustment law In the time window Integral of the absolute value of the internal calculation error: ; Gain adaptive adjustment: ,in, ; Function: Based on the integral of the attitude error over a period of time, dynamically adjust the controller gain; achieve the control objective of fast initial response and stable convergence in the later stage.

[0042] 4) Control Law Structure Pitch channel control input: ; Function: To generate the final control input signal based on the filtering error and variable gain.

[0043] 5) Adaptive parameter update law Define the regression vector: ; Adaptive parameter vector: ; Parameter update law: ; in It is a positive definite diagonal matrix; Parameter update conditions: ; otherwise: ; DSP Discrete Implementation: .

[0044] Function: To estimate system parameters online and compensate for model uncertainties; after discretization, it is used for real-time DSP operation to improve system robustness.

[0045] 3. Position external loop control algorithm 1) High-altitude external loop (50 ms) Height error: ; PI controller: ; Mapped to control input via collective distance channel: ; Function: By using a PI controller to generate collective distance commands based on the altitude error, altitude control is achieved.

[0046] 2) Planar position external loop Position error: , ; PID speed generation command: ; ; The attitude internal loop is used to convert these commands into pitch and roll commands. , ; Function: Generate the desired velocity based on the position error, and then convert it into pitch and roll commands through the attitude inner loop; realize the cascaded control structure and improve trajectory tracking accuracy.

[0047] The sensor module is used to acquire real-time flight status information of the UAV, serving as the fundamental data source for the closed-loop control of the entire flight control system. This module mainly consists of an inertial measurement unit (IMU), a differential global positioning system (DGPS), and a barometer. The IMU integrates a three-axis gyroscope and a three-axis accelerometer to measure the UAV's angular velocity and linear acceleration in the three axes. Integrating the gyroscope's angular velocity yields attitude angle changes, while the accelerometer corrects the attitude by measuring the direction of gravity, thus obtaining stable attitude angle information, including roll, pitch, and yaw angles. Differential GPS is used to acquire high-precision position and velocity information for the UAV. It corrects satellite positioning errors in real time using a ground reference station, enabling the onboard receiver to obtain positioning accuracy and output the UAV's three-dimensional position and ground speed information. The barometer measures ambient air pressure and calculates altitude changes using a standard atmospheric model, obtaining high-resolution altitude data. By fusing barometer and GPS altitude information and using a Kalman filter algorithm to integrate the measurements from the IMU, GPS, and barometer, more accurate and stable UAV state information, including position, velocity, attitude angles, and angular velocity, is obtained, providing reliable state feedback for subsequent flight control algorithms.

[0048] The data processing and calculation module is used to acquire and process sensor data and calculate flight control algorithms. This system adopts a dual-core control architecture of DSP and ARM. The ARM processor is mainly responsible for data acquisition and communication management, while the DSP processor is mainly responsible for the real-time calculation of complex flight control algorithms. The ARM processor acquires sensor data through various communication interfaces, reads attitude and positioning data output from the IMU and differential GPS via a serial port interface, reads barometer altitude information via the I²C bus, and simultaneously receives PPM control signals from the remote controller via a timer capture channel. After obtaining the raw data, the ARM processor preprocesses the data, removing outliers and performing time synchronization. The system performs initial data processing and then writes the processed status data into the dual-port RAM. The DSP processor reads this shared data from the dual-port RAM via the SPI interface and runs the flight control algorithm within it to perform real-time calculations and control decisions on the current flight status of the UAV. The dual-port RAM, as a shared storage unit between the ARM and DSP, allows the two processors to access the same storage space simultaneously without bus conflicts, thus ensuring the real-time performance and reliability of data exchange. Through this dual-core collaborative working mode, the advantages of the ARM in data acquisition and interface management, as well as the advantages of the DSP in high-speed numerical calculation, can be fully utilized, thereby improving the real-time performance and stability of the entire flight control system.

[0049] The adaptive disturbance rejection controller is the core control module of this system. Its main function is to calculate the control quantity based on the UAV's desired flight command and real-time status information, thereby achieving stable control of the UAV's attitude and motion state. This controller adopts an improved model reference adaptive control structure, which constructs a reference model to describe the ideal flight dynamic characteristics. The reference model is usually in the form of a second-order system, with attitude command as input and desired attitude response as output. An attitude tracking error is formed between the UAV's actual attitude and the reference model output. By feeding back this error, the control input can be continuously adjusted so that the actual response gradually approaches the ideal response.

[0050] The variable gain module dynamically adjusts the controller gain based on the integral value of the attitude tracking error within a certain time window. When the system is in the initial stage of control or the attitude error is large, the variable gain module increases the controller gain to enhance the control effect, enabling the system to respond quickly and reduce the error. As the system gradually approaches steady state and the error decreases, the controller gain gradually decreases to reduce the impact of noise on the adaptive parameter update process, thereby suppressing parameter chattering. This ensures a fast system response while improving the stability of parameter convergence, allowing the system to maintain good control performance even in complex environments.

[0051] In a tandem dual-rotor structure, a significant coupling relationship exists between pitch and roll control, meaning the control input of one channel affects the other, increasing the complexity of the control system design. To address this issue, the system first linearizes the nonlinear dynamics model of the UAV near the hovering equilibrium point and establishes the transfer function matrix between the longitudinal and lateral control inputs. Based on this, a dynamic decoupling compensation network is designed to appropriately transform the system inputs, approximating the originally coupled multi-input multi-output system as multiple independent single-input single-output systems in terms of control structure. After decoupling, the pitch and roll channels can be controlled independently, significantly reducing mutual interference between channels and simplifying the subsequent controller design process. This decoupling network operates in real-time in the DSP as a digital filter, processing the control signals before the servo control commands are output.

[0052] The actuator module converts the control signals calculated by the flight controller into actual mechanical actions, thereby changing the UAV's flight state. This module mainly includes a brushless motor, an electronic speed controller (ESC), and multiple digital servos. The brushless motor receives PWM signals from the controller via the ESC to adjust its speed, driving the rotor to generate lift and thrust. The ESC adjusts the motor's current output based on the duty cycle of the input PWM signal, keeping the rotor at a constant or controlled speed. The digital servos manipulate the automatic swashplate in the rotor system, controlling the UAV's attitude by changing the collective pitch and cyclic pitch of the rotors. Pitch control is achieved by changing the cyclic pitch of the front and rear rotors, while roll control is achieved by changing the cyclic pitch in the left and right directions. Multiple servos work together to enable the UAV to perform attitude adjustments, trajectory control, and hovering stabilization.

[0053] The position outer loop controller is used to achieve position and altitude control of the UAV. As the outer loop of the flight control system, this controller takes the desired position, altitude, or velocity command as input and outputs the attitude control command. For altitude control, the system uses a proportional-integral (PI) controller to generate a control quantity based on the error between the current altitude and the target altitude, and adjusts the rotor lift through the collective pitch channel to achieve altitude adjustment. For planar position control, the system calculates the error between the current position and the target position, uses a proportional-integral-derivative (PID) control algorithm to generate a velocity command, and further converts it into pitch and roll attitude commands. When the UAV needs to move forward, the control system generates a certain pitch angle command, causing the UAV to tilt forward, thereby generating horizontal thrust. Through the cascaded control structure between the position outer loop and the attitude inner loop, the UAV can achieve precise positioning and stable flight in space.

[0054] The entire flight control system acquires real-time UAV status information through sensor modules, performs data processing and control calculations via a data processing and calculation module composed of ARM and DSP, and generates control commands through the combined action of an adaptive disturbance rejection controller, a dynamic decoupling unit, and a variable gain mechanism. These commands then drive the rotors and servos via actuators to achieve flight attitude and trajectory control. A closed-loop control structure is formed among the modules, enabling the UAV to stably track the desired flight state even under complex environmental conditions such as wind disturbances, model uncertainties, and sensor noise, thereby improving the system's stability, robustness, and control accuracy.

[0055] Example 2: The operation method of the wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed includes the following steps: Step A1: Obtain real-time status information of the drone through the sensor module; Step A2: In the DSP core of the dual-core controller, run the adaptive disturbance rejection control law. The control law adopts a structure that combines the reference model and the follower model, and uses a variable gain strategy to adjust the error compensation. Step A3: The control law calculates the control quantity for the actuator based on the deviation between the expected command and the state information; Step A4: Drive the drone through the actuator to stably track the desired flight state in an environment with uncertainties such as wind disturbance.

[0056] Example 3: Experiment 1: Comparison of Attitude Tracking Performance under Gust Disturbance Experimental Objective The adaptive disturbance rejection controller with variable gain was verified to have better attitude tracking accuracy and stability compared to traditional MRAC and PID control under wind disturbance conditions.

[0057] Experimental conditions Flight status: hovering in place.

[0058] Wind field conditions: Lateral gust intensity: 4–6 m / s; Gust frequency: 0.3–0.6 Hz.

[0059] Comparison control methods: Traditional PID control; Fixed gain MRAC; The variable gain adaptive disturbance rejection control proposed in this application.

[0060] Experimental indicators Maximum attitude angle error (°); Steady-state root mean square error (RMS) (°); Attitude response overshoot (%).

[0061] Experimental results data Experimental conclusions The results show that under gust disturbance conditions, the variable gain adaptive disturbance rejection control method adopted in this invention can significantly reduce attitude error and overshoot. The attitude tracking accuracy is improved by about 66% compared with traditional PID control and by about 49% compared with fixed gain MRAC, which verifies its robustness and stability in strong wind environment.

[0062] Experiment 2: Verification Experiment of Vertical and Horizontal Dynamic Decoupling Effect Experimental Objective Verify the effect of the dynamic decoupling unit constructed based on the invariance principle on suppressing longitudinal-lateral coupling of a tandem dual-rotor UAV.

[0063] Experimental conditions Flight status: Hovering.

[0064] Operation method: Apply a unit step pitch command (5°); Measure the coupling response of the roll channel.

[0065] Comparison of options: No decoupling control; Static decoupling; The dynamic decoupling unit proposed in this application.

[0066] Experimental indicators Roll maximum coupling angle response (°); Coupling decay time (s).

[0067] Experimental results data Experimental conclusions Experimental results show that the proposed dynamic decoupling unit can effectively suppress the dynamic coupling between the longitudinal and transverse channels, reduce the coupling response amplitude of the roll channel by about 84%, significantly improve the channel independence of the system, and provide a good structural foundation for the design of subsequent adaptive controllers.

[0068] Experiment 3: The Influence of Variable Gain Mechanism on the Convergence of Adaptive Parameters Experimental Objective The effect of the variable gain module on improving the convergence speed of adaptive parameters and the ability to suppress steady-state chattering was verified.

[0069] Experimental conditions Flight status: hovering + minor attitude maneuvers.

[0070] Noise conditions: IMU acceleration noise standard deviation: 0.03 g.

[0071] Comparison of options: Fixed adaptive gain; Variable gain adaptive strategy.

[0072] Experimental indicators Adaptive parameter convergence time : refers to the adaptive parameter vector The time required to adjust from the initial state to a stable range; Steady-state parameter jitter amplitude value (Normalization): refers to the adaptive parameters after the system enters steady state. The amplitude of oscillations around its stable value; in: : Convergence time of adaptive parameters; Steady-state parameter: amplitude value.

[0073] Experimental results data Experimental conclusions Experimental results show that the variable gain mechanism significantly accelerates the convergence speed of the adaptive parameters in the early stage of control, and effectively suppresses parameter chattering caused by noise in the steady state stage, reducing the amplitude of parameter chattering by about 67%, thus solving the problem of insufficient stability of traditional adaptive control in noisy environments.

[0074] Example 4: The specific process for adjusting the speed of a drone: The sensor modules (IMU, differential GPS, barometer) collect the current status (attitude, position, velocity) of the UAV in real time, while the data processing and calculation module is responsible for collecting sensor data.

[0075] The adaptive disturbance rejection controller receives desired commands from the outer loop (position outer loop controller) or remote control station, receives current state information from the ARM core, calculates the deviation between the desired state and the current state, runs an improved model reference adaptive control algorithm, and, in combination with a variable gain module and a dynamic decoupling unit, calculates the total control quantity required to eliminate the deviation and resist wind disturbances.

[0076] The DSP core decomposes the total control input into two parts: Control input decomposition and output. Motor speed control signal: generated by the enhanced PWM module and sent to the electronic speed controller, with the instruction content being "increase speed to XX" or "decrease speed to XX".

[0077] The servo control signal is generated into multiple signals by the PWM module and sent to the digital servo. The instruction is "rotate the forward and backward automatic tilter to XX angle" to change the collective pitch and periodic pitch.

[0078] When the actuator operates, the electronic speed controller adjusts the current supplied to the brushless motor according to the duty cycle of the PWM signal, thereby changing the rotor speed.

[0079] Digital servos drive an automatic swashplate to change the pitch angle of the propeller blades.

[0080] A closed loop is formed, where changes in rotor speed and blade pitch work together to generate new lift and torque on the UAV, thus altering its attitude and position. The sensor module then collects new status information and feeds it back to the controller, initiating a new adjustment cycle.

[0081] Hardware foundation: The brushless motor + electronic speed controller is the component that directly performs speed regulation.

[0082] Command generation: The adaptive disturbance rejection controller in the DSP core calculates the required motor speed command in real time based on the flight status and wind disturbance.

[0083] Control logic: When encountering wind disturbances, the controller will adjust the speed and pitch simultaneously to optimally suppress the disturbances and maintain flight stability.

[0084] Signal format: The adjustment command is output from the PWM module of the flight control computer to the electronic speed controller in the form of a PWM signal.

[0085] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0086] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A wind-driven dynamic control system for the rotational speed of a heavy-duty unmanned aerial vehicle (UAV), characterized in that: include: The sensor module is used to acquire the UAV's status information in real time, including attitude information, position information, and velocity information in the body axis system; The data processing and calculation module is used to collect data from the sensor module and run flight control algorithms. An adaptive disturbance rejection controller is used to receive desired commands and status information and output control quantities. The actuator is used to receive control signals, drive the rotor and control surfaces of the UAV, and suppress the impact of external disturbances on the flight status.

2. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 1, characterized in that: The data processing and calculation module is a dual-core controller based on DSP and ARM. The ARM core is responsible for acquiring data from the sensor module, while the DSP core is used to run flight control algorithms.

3. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 2, characterized in that: The adaptive disturbance rejection controller adopts a structure that combines a reference model and a follower model, and also includes a variable gain module to improve parameter convergence.

4. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 3, characterized in that: The sensor module includes: The IMU (In-Mechanical Unit) attitude system is used to measure the attitude angles, angular rates, and linear acceleration of the UAV. Differential GPS is used to obtain high-precision position and ground speed information for drones; A barometer is used to provide altitude information and integrate it with differential GPS information.

5. The wind-driven dynamic control system for heavy-duty UAV rotation speed according to claim 4, characterized in that: Adaptive disturbance rejection controllers include: The follower model is used to take posture commands as input and output the desired posture response. Reference model, used as input for processed command signals; An error compensator is used to connect the following model and the controlled object; The variable gain module is used to dynamically adjust the gain of the error compensator according to the integral of the attitude tracking error, so as to achieve fast response in the early stage of control and stable convergence in the later stage of control.

6. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 5, characterized in that: The variable gain module is determined by the following formula: ； in, For attitude tracking error, and These are preset parameters.

7. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 6, characterized in that: The adaptive disturbance rejection controller also includes a dynamic decoupling unit, which constructs a decoupling network based on the principle of invariance, enabling independent control of the pitch and roll channels.

8. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 7, characterized in that: The implementing agencies include: Brushless motors and electronic speed controllers are used to drive rotors to maintain a constant or controlled speed; Multiple digital servos are used to manipulate the forward and backward automatic swashplates, which work together to change the collective pitch and cyclic pitch of the rotor to achieve attitude and trajectory control.

9. The wind-driven dynamic control system for heavy-duty unmanned aerial vehicle (UAV) rotation speed according to claim 8, characterized in that: It also includes a position outer loop controller and a ground remote control and telemetry station. The position outer loop controller runs as the outer loop in the DSP core, receives position, speed or altitude commands, and generates inner loop attitude commands based on the deviation from the feedback information from the sensor module, which are then transmitted to the adaptive disturbance rejection controller to form a cascaded control structure. The ground remote control and telemetry station communicates with the data processing and calculation module through a wireless data transmission radio, and is used to send flight mode commands, waypoint tasks and control parameters, receive and display UAV status data and flight trajectory in real time, record flight data, and send remote takeover commands in emergency situations.

10. The operating method of a wind-powered dynamic control system for the rotational speed of a heavy-duty unmanned aerial vehicle (UAV) is characterized by: Operating the wind-driven dynamic control system for heavy-duty UAV rotation speed as described in claim 9 mainly includes the following steps: Step A1: Obtain real-time status information of the drone through the sensor module; Step A2: In the DSP core of the dual-core controller, run the adaptive disturbance rejection control law. The control law adopts a structure that combines the reference model and the follower model, and uses a variable gain strategy to adjust the error compensation. Step A3: The control law calculates the control quantity for the actuator based on the deviation between the expected command and the state information; Step A4: Drive the drone through the actuator to stably track the desired flight state in an environment with uncertainties such as wind disturbance.

Images