Rotor wing-control surface dual-mode wind-resistant attitude stabilizing system and control method based on wind field prediction

Abstract

The invention belongs to the technical field of unmanned aerial vehicle flight control, and particularly relates to a rotor wing-control surface dual-mode wind-resistant attitude stabilizing system and a control method based on wind field prediction, and the method comprises the following steps: S1, collecting flight attitude data, spatial position data and environmental meteorological data of an unmanned aerial vehicle in real time through an airborne multi-source sensor; s2, judging whether the unmanned aerial vehicle enters a strong wind disturbance state or not based on the wind speed data in the environmental meteorological data and the attitude angle change rate in the flight attitude data; s3, when it is judged that the strong wind disturbance state is entered, the dynamic target balance attitude is calculated according to the current environment wind speed, wind direction and flight mode; according to the invention, the technical problems of inaccurate strong wind identification, insufficient dual-mode cooperation, high attitude return energy consumption, lack of safety protection and the like in the wind-resistant attitude stabilization control of the existing composite wing unmanned aerial vehicle can be effectively solved, and accurate attitude stabilization, energy consumption optimization and safety protection in a complex wind field are realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of unmanned aerial vehicle flight control, and particularly relates to a rotor-surface dual-mode wind-resistant stable attitude system and a control method based on wind field prediction. BACKGROUND

[0002] The blended wing unmanned aerial vehicle combines the vertical take-off and landing capability of the multi-rotor unmanned aerial vehicle and the high-speed cruising advantage of the fixed-wing unmanned aerial vehicle, and is widely applied in surveying, inspection, emergency rescue and the like. However, when operating outdoors, the unmanned aerial vehicle often faces complex wind field disturbance such as strong wind and gust, and is prone to attitude out-of-control, flight path deviation and even crash risk.

[0003] The existing wind-resistant stable attitude control technology has many deficiencies: firstly, the strong wind state recognition mechanism is single, and only relies on the wind speed threshold or a single attitude change index, which is prone to false triggering (excessive reaction in light wind) or missed triggering (not responding in time in strong wind); secondly, the control mode adaptability is poor, and a single rotor or surface control logic is mostly used, without fully exerting the advantages of dual-mode cooperation, and the wind-resistant capability is weak in the transition mode; thirdly, the attitude return strategy is rigid, with a fixed horizontal attitude as the target, without considering the moment balance under the wind field, resulting in high energy consumption, response lag and obvious overshoot; and fourthly, there is a lack of full-link safety protection mechanism, and the response to risks such as power load overrun and energy shortage is insufficient, which is prone to equipment damage. SUMMARY

[0004] The purpose of the present application is to provide a rotor-surface dual-mode wind-resistant stable attitude system and a control method based on wind field prediction, which can effectively solve the technical problems of inaccurate strong wind recognition, insufficient dual-mode cooperation, high energy consumption in attitude return and lack of safety protection in the existing composite wing unmanned aerial vehicle wind-resistant stable attitude control, and realize precise stable attitude, energy consumption optimization and safety protection in complex wind field.

[0005] The technical solution adopted by the present application is as follows: A rotor-surface dual-mode wind-resistant stable attitude control method based on wind field prediction, comprising the following steps: S1: collecting the flight attitude data, spatial position data and environmental meteorological data of the unmanned aerial vehicle in real time through the on-board multi-source sensor; S2: judging whether the unmanned aerial vehicle enters a strong wind disturbance state based on the wind speed data in the environmental meteorological data and the attitude angle change rate in the flight attitude data; S3: when it is judged that the strong wind disturbance state is entered, calculating a dynamic target balance attitude according to the current environmental wind speed, wind direction and flight mode, the dynamic target balance attitude being a stable attitude that can make the unmanned aerial vehicle maintain the optimal energy in the current wind field; S4: generating an attitude return control instruction based on the deviation between the dynamic target balance attitude and the current actual attitude; S5: distributing the attitude back-to-center control instruction to the power system and control surface actuator of the UAV to realize attitude back-to-center control.

[0006] The S1 on-board multi-source sensor includes an inertial measurement unit, a GPS module and a micro anemometer; the sampling frequency of the inertial measurement unit is not less than 100 Hz, and the attitude angle measurement accuracy is ≤±0.1°; the positioning accuracy of the GPS module is ≤1 m, and the speed measurement accuracy is ≤0.1 m / s.

[0007] The specific method for determining whether the UAV enters a strong wind disturbance state in S2 includes: comparing the current environmental wind speed with a preset strong wind threshold value, and triggering a strong wind state flag when the wind speed exceeds the strong wind threshold value; calculating the attitude angle change rate of the current flight attitude data, and triggering a disturbance state flag when the attitude angle change rate exceeds a preset disturbance threshold value; when the strong wind state flag and the disturbance state flag are triggered at the same time, it is determined that the strong wind disturbance state is entered.

[0008] The calculation method of the dynamic target balance attitude in S3 includes: based on the current environmental wind speed and wind direction, the aerodynamic force and aerodynamic moment acting on the UAV body are estimated through the aerodynamic model; according to the aerodynamic moment, the balance attitude angle required to offset the wind disturbance is calculated, including the roll balance angle, the pitch balance angle and the yaw balance angle; combining the current flight mode of the UAV, the balance attitude angle is corrected to obtain the dynamic target balance attitude.

[0009] The flight mode correction includes: multi-rotor mode correction: limiting the amplitude of the roll balance angle and the pitch balance angle, smoothing the attitude change rate through a first-order low-pass filter, weakening the yaw balance angle and maintaining the original heading; fixed-wing mode correction: coupling the roll balance angle and the yaw balance angle to form a coordinated turning attitude, dynamically compensating the pitch angle according to the current airspeed, and setting the maximum lift angle of the pitch angle to change with the airspeed to realize stall boundary protection.

[0010] The generation of the attitude back-to-center control instruction in step S4 adopts a hierarchical control strategy: outer loop control layer: taking the dynamic target balance attitude as the set value, a fuzzy self-adaptive controller is used to generate the expected angular rate command; inner loop control layer: based on the expected angular rate command, a PID controller is used for rapid angular rate stability control to generate the final control instruction.

[0011] The method adopts different control strategies according to different flight modes of the compound wing unmanned aerial vehicle: When in the multi-rotor mode, roll and pitch attitude control is realized by adjusting the speed difference of the multiple rotors; When in the fixed wing mode, attitude control is realized by coordinating the control of the rudder deflection angle and the aerodynamic thrust.

[0012] The method further includes safety monitoring, real-time monitoring of the load state and energy consumption of the power system of the unmanned aerial vehicle, and when it is detected that the load of the power system exceeds a safety threshold or the energy consumption exceeds an alert value, a three-level degradation protection strategy is started.

[0013] The three-level degradation protection strategy includes: the first level protection is load buffering, relaxing the attitude allowable deviation and lowering the upper limit of power output; The second level protection is emergency return, planning a path with minimum wind component and matching a safe landing point; The third level protection is ultimate protection, outputting the maximum authority attitude adjustment instruction, and triggering the parachute device after failure.

[0014] A rotor-rudder dual-mode wind-resistant stable attitude system, comprising: A sensor data acquisition module acquires real-time data of the attitude angle (roll angle, pitch angle, yaw angle), angular rate, position, height, and environmental wind speed and direction of the unmanned aerial vehicle through an IMU (inertial measurement unit), GPS, barometer, and onboard anemometer; A strong wind disturbance judgment module accurately identifies the strong wind disturbance state by using a dual judgment mechanism of wind speed threshold and attitude change rate. When the wind speed exceeds the set threshold and the attitude angle change rate is abnormal, it is determined that the strong wind disturbance state is entered; A dynamic target attitude calculation module calculates the "dynamic target balance attitude" in real time based on the current wind field conditions and the flight mode. This attitude is not an absolute horizontal zero-degree attitude, but an optimal attitude that can keep the unmanned aerial vehicle stable and consume the least energy in the current wind field; An intelligent return-to-center control module adopts a double-layer control architecture of outer loop-inner loop. The outer loop generates the expected angular rate command based on the dynamic target attitude, and the inner loop realizes fast angular rate tracking control; A safety monitoring protection module monitors the power system state in real time, automatically degrades the control requirements or starts the emergency program when the load is too large, and ensures flight safety; A flight mode adaptive module adopts different control strategies and parameters in the multi-rotor mode and the fixed wing mode according to the characteristics of the compound wing.

[0015] A computer readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the control method described above.

[0016] The technical effects achieved by the present application are as follows: The present application adopts a wind speed threshold and attitude change rate dual coordination determination mechanism, completely solves the problem of easy false triggering or missed triggering of traditional single indicators, improves the strong wind state recognition accuracy compared with traditional methods, and ensures the timeliness and reliability of the control system response.

[0017] The dynamic target balance attitude adopted by the present application combines the double-layer control architecture of outer loop fuzzy self-adaptation + inner loop PID, which adapts to multi-rotor and fixed-wing modes, and also covers mode transition scenarios. Effectively reduce the attitude overshoot, shorten the attitude recovery time, reduce the attitude fluctuation amplitude in the transition mode, and effectively suppress the attitude oscillation and track deviation under strong wind.

[0018] The present application uses the principle of moment balance to use gravitational moment and aerodynamic restoring moment to assist wind resistance, rather than simply relying on the power system to offset wind disturbance, reduces the energy consumption of the wind resistance process, improves the endurance of the unmanned aerial vehicle, and is especially suitable for long-time outdoor operation scenarios.

[0019] The three-level degradation protection strategy adopted by the present application covers the full risk chain of power load overrun, energy shortage, and attitude instability, reduces the incidence of crash accidents, reduces equipment loss, and triggers the response time of parachute landing in emergency state ≤300ms. Independent relay control ensures the reliability of parachute landing, and at the same time, real-time broadcast of fault information provides support for subsequent recovery and fault troubleshooting.

[0020] The present application fully utilizes the dual-mode synergy advantages of fast response of rotors and high efficiency of rudder surfaces, solves the pain point of weak wind resistance ability of traditional single-mode control of compound wing unmanned aerial vehicles, and expands the maximum wind speed from traditional 12m / s to above 25m / s, improves the wind resistance level by one level, and can adapt to complex outdoor wind field operation requirements.

[0021] The present application customizes differentiated correction strategies and control parameters for different flight modes, realizes mode adaptive switching under strong wind without manual intervention, reduces the operation threshold, and transmits control instructions through CAN bus or high-speed PWM signal, with a response delay ≤10ms, which improves the action accuracy of the execution mechanism. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 It is the control method flowchart of the present application; Figure 2 It is the control system architecture block diagram of the present application; Figure 3 It is the attitude recovery control method flowchart of the present application; Figure 4 It is the dual-mode wind resistance and attitude stabilization system block diagram of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose and advantages of the present application more clear and apparent, the present application will be specifically described below in conjunction with embodiments. It should be understood that the following description is merely used to describe one or several specific embodiments of the present application and does not strictly limit the scope of protection specifically requested by the present application.

[0024] As shown in Figure 1 and Figure 3 , a rotor-surface dual-mode anti-wind stability control method based on wind field prediction comprises the following steps: S1: Real-time acquisition of flight attitude data, spatial position data and environmental meteorological data of the unmanned aerial vehicle through an on-board multi-source sensor, to provide accurate input for subsequent strong wind determination and control instruction generation; The attitude angle and angular rate of the unmanned aerial vehicle are acquired through an inertial measurement unit, with a sampling frequency not less than 100 Hz and an attitude angle measurement accuracy controlled within ±0.1°; The spatial position coordinates (longitude, latitude, altitude) and flight speed of the unmanned aerial vehicle are acquired through a GPS module, with a positioning accuracy ≤1 m and a speed measurement accuracy ≤0.1 m / s; The wind speed and direction of the environment are acquired through a miniature anemometer.

[0025] S2: Based on the wind speed data in the environmental meteorological data and the attitude angle change rate in the flight attitude data, it is determined whether the unmanned aerial vehicle enters a strong wind disturbance state; A strong wind threshold is preset, combined with the maximum wind resistance level of the unmanned aerial vehicle, the wing load and other aerodynamic parameters, and a strong wind threshold is finally calibrated through regression analysis of ground wind tunnel test and outdoor test flight data, taking into account flight safety and control efficiency; The actual environmental wind speed is compared with the strong wind threshold, and the strong wind state flag is triggered when the wind speed exceeds the strong wind threshold; A disturbance threshold is preset and calibrated through ground vibration test and strong wind test, to calibrate a disturbance threshold that can effectively represent the beginning of attitude loss of control or severe oscillation; The attitude angle change rate of the actual flight attitude data is calculated in real time, and when any one of the attitude angle change rates exceeds the preset disturbance threshold, the disturbance state flag is triggered; Only when the strong wind state flag and the disturbance state flag are triggered at the same time, it is determined that the unmanned aerial vehicle has entered a state. Once it is determined to enter the state, the control system immediately interrupts the normal flight control mode and activates the subsequent intelligent attitude return process. If the above two flags are not met at the same time, it is determined that the current wind disturbance is within a controllable range, and the system returns to step S1 to continue monitoring and maintaining the original normal flight control mode. Through this double and cooperative judgment mechanism, the accuracy and reliability of state recognition are greatly improved, ensuring that the control system neither overreacts in light wind nor is timely and effectively activated under real strong wind threat.

[0026] S3: When it is determined to enter the strong wind disturbance state, Taking the current collected environmental wind speed and direction as input, the pre-stored UAV aerodynamic model is called to match the current flight state with the aerodynamic model, and the aerodynamic force and moment generated by the wind field acting on the UAV body are estimated in real time; According to the estimated aerodynamic moment, based on the rigid body dynamics and torque balance principle, the basic balance attitude angle required to statically offset the wind disturbance is calculated, which provides a reliable theoretical benchmark for subsequent flight mode correction. The specific calculation process is as follows: A1. Establish torque balance relationship: In the balanced state, the wind disturbance aerodynamic moment, the gravity moment and the control moment provided by the power system together constitute the balance; the goal of the calculation is to find an optimal attitude, so that the gravity moment can maximize the wind disturbance aerodynamic moment, thereby significantly reducing the control moment required by the power system, and achieving the optimal energy consumption of the stable state; A2. Generation mechanism and application of gravity moment: The gravity moment is the core of passive balance; when the UAV generates a roll or pitch attitude angle, the gravity line relative to the aerodynamic center will generate a restoring moment; the size of this gravity moment is approximately proportional to the mass of the UAV, the height of the center of gravity and the sine value of the attitude angle; the system actively and accurately controls the roll angle and pitch angle to generate a gravity restoring moment which is opposite in direction and similar in size to the wind disturbance aerodynamic moment, thereby forming a balance; A3. Solve the balance attitude angle of each axis: Roll and pitch balance angle: the system solves the nonlinear torque balance equation set to solve the roll balance angle and pitch balance angle; the quantitative relationship is: the required balance angle mainly depends on the size of the wind disturbance aerodynamic moment, and is inversely proportional to the weight and height of the center of gravity of the UAV; that is, the larger the wind disturbance moment, the larger the required balance angle; the heavier the UAV or the higher the center of gravity, the smaller the balance angle required to generate the same restoring moment; A4. Yaw balance angle: yaw balance is mainly achieved through aerodynamic coordination; to balance the wind-induced yaw moment, the system calculates a yaw balance angle (i.e. sideslip angle), which is to make the UAV nose deflect appropriately to form an "upwind" or "slip" attitude; in this way, a aerodynamic restoring moment opposite in direction to the wind-induced yaw moment can be generated on the vertical tail and other aerodynamic surfaces, thereby achieving balance on the yaw axis; the size of the angle is proportional to the wind-induced yaw moment and inversely proportional to the directional static stability derivative of the UAV; A5. Through the above specific physical modeling and engineering calculation, a set of accurate initial balance attitude angles are determined; According to the current flight mode, the initial balance attitude angle is corrected, and the specific correction strategy is as follows: Multi-rotor mode correction strategy Attitude angle amplitude limiting, strict absolute value upper limit is imposed on the roll balance angle and the pitch balance angle; Attitude change rate smoothing filtering, the corrected target attitude angle is filtered through a first-order low-pass filter, and the filter time constant is set according to the dynamic response speed of the rotor and the moment of inertia of the body; Yaw angle decoupling processing, in multi-rotor mode, the yaw balance angle is weakened; The roll angle and the pitch angle after amplitude limiting and smoothing filtering, and the yaw angle maintaining the original heading, jointly constitute the dynamic target balance attitude in multi-rotor mode; Fixed-wing mode correction strategy Roll-yaw coupling coordination, the roll balance angle and the yaw balance angle are coupled to calculate a coordinated turn attitude; according to the expected track correction amount and the roll balance angle, a target roll angle is determined, and then according to the target roll angle and the airspeed, the corresponding target yaw angle and airspeed are calculated through the coordinated turn equation; Pitch angle airspeed compensation, the pitch balance angle is dynamically corrected according to the current airspeed; when flying against the wind, the airspeed increases and the lift increases, so the pitch angle needs to be reduced to prevent climbing; when flying with the wind, the airspeed decreases and the lift decreases, so the pitch angle needs to be increased to prevent height loss; Stall boundary protection, a negative upper limit is set for the final calculated pitch angle, i.e. the maximum lift angle, which dynamically changes with the airspeed; The roll angle, the pitch angle, and the yaw angle after coordinated coupling, airspeed compensation, and protection amplitude limiting jointly constitute the dynamic target balance attitude in fixed-wing mode.

[0027] S4: based on the deviation between the dynamic target balance attitude and the current actual attitude, generate attitude return control instructions; the generation of attitude return control instructions adopts a double-layer control architecture of outer loop-inner loop; The outer loop generates expected angular rate instructions based on the dynamic target attitude, Outer loop control layer The dynamic target balance attitude is used as the set value, and a fuzzy adaptive controller is adopted; the difference between the dynamic target balance attitude and the current actual attitude in each axis and the differential of the attitude deviation are used as input variables, and the input variables are converted into fuzzy language variables according to the preset membership function; the fuzzy reasoning is based on the fuzzy rule base summarized from expert experience and a large number of simulation tests, which enables the controller to intelligently adjust the control strategy according to the size and trend of the deviation; The fuzzy quantity output by the fuzzy reasoning is de-fuzzied through the barycentric method and converted back to the accurate expected angular rate instruction When the attitude deviation is large, the controller outputs a larger angular rate command to achieve fast response; when approaching the target attitude, it outputs a small and soft command to effectively suppress overshoot and oscillation, and achieve smooth return without impact; The inner loop realizes fast angular rate tracking control; The inner loop control layer: PID controller is adopted, based on the expected angular rate command input, the controller calculates the required control torque through the linear combination of the proportional, integral and differential three links according to the angular rate deviation; for the dramatic angular motion of the unmanned aerial vehicle in strong wind, the differential term is usually optimized, the differential pre-act or incomplete differential strategy is added to suppress the measurement noise amplification and improve the system stability; The output of the controller is the final control command, which is distributed according to the current flight mode: The command of the multi-rotor mode is converted into the thrust difference of multiple motors, and the motor speed is adjusted through the electronic speed controller (ESC) to generate the control torque; The command of the fixed-wing mode is converted into the deflection angle command of each aerodynamic control surface (aileron, elevator and rudder), and the deflection of the control surface is driven by the servo to generate the aerodynamic control torque; Based on the expected angular rate command, PID controller is used for fast angular rate stable control to generate the final control command.

[0028] S5: Distribute the attitude return control command to the power system and control surface actuator of the unmanned aerial vehicle to realize attitude return control; the communication protocol uses high-reliability CAN bus or high-speed PWM signal, and the control command is encoded into the standard signal format recognizable by the actuator; for the power system, it is converted into the throttle amount or motor target speed; for the control surface, it is converted into the target deflection angle of the servo; According to the current flight mode, the command is distributed to different actuator combinations; In the multi-rotor mode: the roll, pitch and yaw torque commands and the total lift command are solved into the independent speed command of each motor through the control distribution matrix, and the recovery torque is generated by adjusting the speed difference between the rotors to drive the attitude of the unmanned aerial vehicle to return to the center; In the fixed-wing mode: the control command is distributed to each aerodynamic control surface, the differential deflection of the aileron is realized, the deflection of the elevator is realized, the deflection of the rudder is realized, and the sideslip coordinated turn is realized in coordination with the aileron; After the actuator drives the unmanned aerial vehicle body to produce corresponding angular motion and linear motion, a closed loop feedback is formed: State update: the onboard IMU, GPS and other sensors sense the new attitude and angular rate generated by the actuator action in real time; Feedback loop: these new state data are immediately fed back to S1 through multiple sensors; Continuous correction, the controller compares the new actual attitude with the dynamic target balance attitude again, generates new control instructions to correct the residual deviation still existing after the previous control.

[0029] The method also includes safety monitoring, real-time monitoring of the load state and energy consumption of the unmanned aerial vehicle power system, and starting a three-level degradation protection strategy based on the monitoring results; Primary protection: load buffer protection As the first line of defense, it aims to reduce power load by flexibly adjusting control strategy to avoid entering extreme working conditions. Its trigger conditions and execution logic are as follows: Trigger condition: meet any of the following conditions for 500ms (to avoid false triggering due to transient disturbance): ① motor current exceeds 80% rated current but does not reach 110% limit value; ② motor temperature exceeds 85℃ warning threshold but does not reach 95℃ limit value.

[0030] Execution process: Dynamic relaxation of control accuracy, the flight controller automatically adjusts the allowed deviation of the dynamic target balance attitude from ±0.5° to ±1°, reducing the rigidity requirement of attitude tracking and reducing the power loss caused by frequent adjustment; Intelligent limiting of power output, issuing "energy consumption priority" control instructions to the fuzzy adaptive controller, dynamically adjusting the upper limit of power output according to the current load exceeding amplitude - when the current / temperature is 1.0-1.05 times the warning threshold, the output torque upper limit is reduced by 15%; when it is 1.05-1.1 times, it is reduced by 20%, ensuring that the load reduction amplitude matches the load state; State closed loop verification, after executing the load reduction strategy, continuously monitor the motor current and temperature, if the parameters fall below the warning threshold within 5 seconds, maintain the adjusted control parameters; if they are still in the exceeding range or show an upward trend, trigger the second level of protection immediately.

[0031] Secondary protection: emergency return protection When the primary protection cannot effectively alleviate the load pressure, or the energy system is at risk, start the emergency return program to plan the return path with the goal of "low energy consumption, high safety", the specific scheme is: Trigger condition: ① motor current / temperature still exceeds the standard after primary protection is executed for 5 seconds; ② battery remaining power continuously below 20% warning value for 3 seconds, or power consumption per unit time exceeds the reference value by 50% and lasts for 2 seconds; ③ GPS signal is normal (to ensure that the path planning is effective).

[0032] Execution process: Optimal return path planning, the flight controller calls the pre-stored electronic map containing terrain elevation, no-fly zone data, and combines the wind field data collected by the on-board anemometer in real time, to calculate the optimal return path by A Algorithm planning "minimum upwind component" path - screening the optimal solution of the candidate path with upwind segment proportion ≤20%, altitude margin ≥50m (avoiding obstacles), and calculating the minimum energy required for return to ensure that the battery power meets the flight requirements; Safe landing point matching, preferentially matching the pre-stored three primary safety areas, meeting the area ≥10m×10m, no obstruction, and GPS signal strength ≥-85dBm. If the primary area is not available, such as being temporarily occupied, the secondary safety area is automatically matched, with an area ≥8m×8m, and the ground control station is sent alarm information containing the landing point coordinates, area photos, and recommended landing attitude; Return attitude control optimization, switching to "return exclusive control parameters" - increasing the cruise altitude by 20% in fixed-wing mode to enhance wind resistance, and reducing the hover height to 50m in multi-rotor mode to reduce wind disturbance; At the same time, the battery energy consumption is allocated as "flight energy consumption 60% + landing redundancy 40%", to avoid insufficient power during landing.

[0033] Three levels of protection: ultimate protection For attitude instability precursors, start the highest level of protection measures to minimize crash losses. The trigger logic and execution scheme are as follows: Trigger condition: roll angle, pitch angle, or yaw angle deviation continuously exceeds ±10° for 1 second, and is verified by IMU and GPS data fusion (attitude anomaly and position deviation trend are consistent), to exclude false positives caused by sensor failure.

[0034] Execution process: Emergency attitude correction, the flight controller immediately outputs the maximum authority attitude adjustment instruction (the rudder deflection reaches 80% of the limit value, and the rotor speed difference reaches 90% of the maximum allowed value), trying to suppress attitude deterioration, with a correction time limit of 300ms; Parachute trigger, if the attitude deviation is not reduced to within ±5° within 300ms, it is determined to be out of control risk, and the flight controller triggers the parachute device directly through an independent relay contact (not affected by the main control loop), while cutting off the power supply to the power system, to avoid secondary injury caused by high-speed rotating propellers; Emergency state broadcast, after triggering the parachute, an emergency message is sent to the ground control station at a frequency of 10Hz, including the current GPS coordinates, attitude data, battery remaining capacity, and fault code, until the signal is interrupted or the UAV lands.

[0035] As shown in Figure 2 and Figure 4 , a rotor-rudder dual-mode wind-resistant attitude stabilization system includes: The sensor data acquisition module collects real-time data on the UAV's attitude angles, roll angle, pitch angle, yaw angle, angular rate, position, altitude, and ambient wind speed and direction through IMU (Inertial Measurement Unit), GPS, barometer, and airborne anemometer. The strong wind disturbance judgment module integrates wind speed threshold and attitude change rate as dual judgment mechanisms to accurately identify strong wind disturbance state; when the wind speed exceeds the set threshold and the attitude angle change rate is abnormal, it is determined that the strong wind disturbance state has been entered. The dynamic target attitude calculation module calculates the "dynamic target equilibrium attitude" in real time based on the current wind field conditions and flight mode. This attitude is not an absolute horizontal zero-degree attitude, but the optimal attitude that enables the UAV to maintain stability and consumes the least energy in the current wind field. The intelligent return-to-center control module adopts a two-layer control architecture of outer loop and inner loop; the outer loop generates the desired angular rate command based on the dynamic target attitude, and the inner loop realizes fast angular rate tracking control. The safety monitoring and protection module monitors the power system status in real time and automatically degrades control requirements or initiates emergency procedures when the load is too high to ensure flight safety. The flight mode adaptive module employs different control strategies and parameters in multi-rotor and fixed-wing modes, taking into account the characteristics of the compound wing.

[0036] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the control method described above.

[0037] Example 1: Strong Wind Disturbance Control in Fixed-Fly Mode A compound-wing UAV encountered strong crosswinds while cruising in fixed-wing mode. The onboard sensors collected data in real time: the inertial measurement unit (IMU) detected that the roll angle change rate reached 10° / s within 0.2 seconds, the preset disturbance threshold was 5° / s, the GPS module positioning accuracy was ≤1m, the speed measurement accuracy was ≤0.1m / s, and the miniature anemometer measured the wind speed at 18m / s (the preset strong wind threshold was 12m / s). Based on the wind speed exceeding the threshold and the abnormal roll angle change rate, the system made a dual judgment that it had entered a strong wind disturbance state, immediately interrupted the normal cruise mode, and activated the intelligent attitude return process. The dynamic target attitude calculation module calls the pre-stored aerodynamic model according to the 18 m / s crosswind condition, and calculates the optimal balance attitude through the rigid body dynamics and torque balance principle; in the calculation, the aerodynamic moment of the crosswind on the aircraft is considered, and based on the roll-yaw coupling coordination strategy: the roll angle + 5° can produce a gravity moment to offset the roll, and the yaw angle - 3° can balance the yaw moment through the aerodynamic effect of the rudder; at the same time, the airspeed compensation module dynamically adjusts the pitch angle to -1° according to the current airspeed to prevent excessive lift caused by headwind; the stall boundary protection module sets the upper limit of the pitch angle to +10° to avoid the risk of stall; the final dynamic target attitude is determined as roll angle + 5°, pitch angle -1°, yaw angle -3°, which can minimize energy consumption and stabilize the aircraft using aerodynamic restoring moment; The intelligent back-to-center control module adopts a double-layer architecture of outer loop-inner loop: The outer loop fuzzy adaptive controller takes the dynamic target attitude as the set value, inputs the roll angle deviation (the difference between the current actual roll angle and the target + 5°) and its derivative, and obtains a smooth expected angular rate command through the fuzzy rule base (such as "large deviation, large output angular rate"); The inner loop PID controller receives the angular rate command, optimizes the derivative term for fixed-wing mode (adds an incomplete derivative strategy), outputs the control moment and distributes it to the actuators: aileron differential deflection realizes roll control, elevator deflection realizes pitch control, rudder deflection realizes yaw control, and the engine thrust is appropriately increased to maintain airspeed; The safety monitoring module monitors the actuator load and battery status in real time, and maintains the current strategy as long as they are within the safe range; compared with traditional PID control, the unmanned aerial vehicle quickly stabilizes to the target attitude in strong wind, with a 40% reduction in attitude overshoot, a 35% reduction in recovery time, and a 20% reduction in energy consumption.

[0038] Embodiment 2: strong wind disturbance control in transition mode When a compound wing unmanned aerial vehicle is transitioning from multi-rotor mode to fixed-wing mode, it encounters sudden gusts; onboard multi-source sensor data: IMU detects that the pitch angle rate changes to 12° / s within 0.4 seconds, and the roll angle rate changes to 9° / s, the pre-set disturbance threshold is 5° / s, the GPS measures the horizontal speed to accelerate to 15 m / s, and the miniature anemometer shows that the wind speed is 20 m / s, and the pre-set strong wind threshold is 12 m / s; the system determines that it enters the strong wind disturbance state according to the wind speed exceeding the threshold and multiple abnormal attitude change rates, immediately terminates the transition process, and activates the intelligent wind-resistant control; The dynamic target attitude calculation module combines the aerodynamic characteristics of the transition mode, the rotor and the rudder surface cooperate, and calculates the initial balance attitude through the aerodynamic model: pitch angle + 4°, roll angle -3°, yaw angle + 2°; in the correction strategy: Multirotor component: absolute value clipping on roll and pitch angle (±8°), smoothed by a first order low pass filter with time constant 0.1 s; Fixed wing component: roll-yaw coupling coordination, yaw angle +2° calculated from roll angle -3° to form coordinated turn; pitch angle airspeed compensation adjusted to +3° according to current airspeed to prevent downwind lift loss; Final dynamic target attitude: pitch angle +3°, roll angle -2°, yaw angle +2°, which takes full advantage of the fast response of the rotor and the aerodynamic efficiency of the rudder; The outer loop fuzzy adaptive controller of the intelligent centering control module takes the target attitude as input to generate angular rate commands; the inner loop controller mixes and distributes commands: in multirotor mode, adjust motor speed difference to generate restoring moments, in fixed wing mode, control aileron, elevator and rudder deflection; the safety monitoring module monitors motor and servo load, although there is a slight rise but not over limit, maintain control strategy; the UAV smoothly completes mode transition in gust, attitude fluctuation amplitude is reduced by 50% compared with traditional method, transition time is reduced by 30%.

[0039] Example 3: Wind-resistant control triggered by safety monitoring to trigger degradation protection A compound wing UAV was in multirotor mode hovering when it encountered sustained strong wind; sensor data: IMU detected that the pitch angle rate reached 15° / s within 0.3 seconds, the preset disturbance threshold was 5° / s, the anemometer measured the wind speed as 25 m / s, the preset strong wind threshold was 12 m / s, the system determined that it entered the strong wind disturbance state; the dynamic target attitude calculation module calculated the optimal balance attitude as pitch angle +6°, roll angle -4°, and the intelligent centering control module started motor speed adjustment; At the same time, the safety monitoring module detected that the motor current exceeded 80% of the rated current for 500 ms, reaching 105% of the rated value, and the motor temperature rose to 88°C, triggering level one protection: Dynamic relaxation of control accuracy, attitude allowed deviation from ±0.5° to ±1°; Power output intelligent clipping, fuzzy adaptive controller switched to "energy priority" mode, output torque upper limit down by 20%; Within 5 seconds after execution, the motor current fell back to the safe range, but the battery power remained below the 20% warning value for 3 seconds, triggering level two protection, emergency return protection: Optimal return path planning: based on electronic map and wind field data, through A Algorithm selects paths with windward segment ratio ≤20% and altitude margin ≥50 m; Safe landing point matching: preferentially select pre-stored level one safe area with an area of ≥10m×10m, and send coordinates to the ground station; Return attitude control optimization: the multi-rotor mode hover height is reduced to 50m, and the battery energy consumption distribution is 60% for flight and 40% for landing redundancy; During the return, the IMU and GPS data fusion shows that the pitch angle deviation continues to exceed ±10° for 1 second, triggering the third level protection, and the ultimate protection: Emergency attitude correction: output maximum authority command, rotor speed difference reaches the limit value 90%, try to correct within 300ms; Parachute device trigger: after correction invalid, trigger parachute and cut off power through independent relay; The unmanned aerial vehicle safely lands, and the ground station receives emergency messages containing GPS coordinates and fault codes; the entire process avoids crashing, and compared with traditional methods, the loss is reduced by 70%.

[0040] The above is only the preferred embodiment of the present application, it should be noted that for those skilled in the art, without departing from the principles of the present application, can make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application. The structures, devices and operation methods not specifically described and explained in the present application, such as no special description and limitation, are implemented according to the conventional means in the art.

Claims

1. A rotor-surface dual-mode anti-wind stability control method based on wind field prediction, characterized in that, The method comprises the following steps: S1: Real-time collection of flight attitude data, spatial position data and environmental meteorological data of the UAV through an on-board multi-source sensor; S2: Judgment of whether the UAV enters a strong wind disturbance state based on wind speed data in the environmental meteorological data and attitude angle change rate in the flight attitude data; S3: When it is judged that the strong wind disturbance state is entered, a dynamic target balance attitude is calculated according to the current environmental wind speed, wind direction and flight mode, the dynamic target balance attitude being a stable attitude that can make the UAV maintain optimal energy in the current wind field; S4: Generation of a posture return-to-center control instruction based on the deviation of the dynamic target balance attitude from the current actual attitude; S5: Distribution of the posture return-to-center control instruction to the power system and the control surface actuator of the UAV to realize posture return-to-center control.

2. The control method according to claim 1, characterized by: The on-board multi-source sensor of S1 comprises an inertial measurement unit, a GPS module and a micro anemometer; the sampling frequency of the inertial measurement unit is not less than 100 Hz, and the attitude angle measurement accuracy is ≤±0.1°; the positioning accuracy of the GPS module is ≤1 m, and the speed measurement accuracy is ≤0.1 m / s.

3. The control method according to claim 1, characterized by: The specific method for judging whether the UAV enters a strong wind disturbance state in S2 comprises: Comparison of the current environmental wind speed with a preset strong wind threshold value, and triggering of a strong wind state flag when the wind speed exceeds the strong wind threshold value; Calculation of the attitude angle change rate of the current flight attitude data, and triggering of a disturbance state flag when the attitude angle change rate exceeds a preset disturbance threshold value; When the strong wind state flag and the disturbance state flag are triggered at the same time, it is judged that the strong wind disturbance state is entered.

4. The control method according to claim 1, characterized by: The calculation method of the dynamic target balance attitude in S3 comprises: Estimation of the aerodynamic force and the aerodynamic moment acting on the UAV body through an aerodynamic model based on the current environmental wind speed and wind direction; Calculation of the balance attitude angle required to offset the wind disturbance, including the roll balance angle, the pitch balance angle and the yaw balance angle, according to the aerodynamic moment; Correction of the balance attitude angle in combination with the current flight mode of the UAV to obtain the dynamic target balance attitude.

5. The control method according to claim 1, characterized by: The flight mode correction comprises: Multi-rotor mode correction: amplitude limiting of the roll balance angle and the pitch balance angle, smoothing of the attitude change rate through a first-order low-pass filter, weakening of the yaw balance angle and maintenance of the original heading; Fixed-wing mode correction: coupling solution of the roll balance angle and the yaw balance angle to form a coordinated turning attitude, dynamic compensation of the pitch angle according to the current airspeed, and setting of a maximum lift angle of the pitch angle that changes with the airspeed to realize stall boundary protection.

6. The control method according to claim 1, characterized by: The posture return-to-center control instruction in S4 adopts a hierarchical control strategy: Outer loop control layer: taking the dynamic target balance attitude as a set value, a fuzzy self-adaptive controller is used to generate an expected angular rate instruction; Inner loop control layer: based on the expected angular rate instruction, a PID controller is used for rapid angular rate stable control to generate a final control instruction.

7. The control method according to claim 1, characterized by: The method adopts different control strategies according to different flight modes of the compound wing UAV: When in the multi-rotor mode, the roll and pitch attitude control is realized by adjusting the rotational speed difference of multiple rotors; When in the fixed-wing mode, the attitude control is realized by coordinately controlling the deflection angle of the rudder and the aerodynamic thrust.

8. The control method according to claim 1, characterized by: The method also includes safety monitoring, real-time monitoring of the load state and energy consumption of the unmanned aerial vehicle power system, and starting a three-level degradation protection strategy when detecting that the power system load exceeds a safety threshold or the energy consumption exceeds an alert value; The three-level degradation protection strategy includes: the first level protection is load buffering, relaxing the attitude allowed deviation and down-regulating the upper limit of power output; The second level protection is emergency return, planning a path with minimum wind component and matching a safe landing point; The third level protection is ultimate protection, outputting maximum authority attitude adjustment instructions, and triggering parachute devices after failure.

9. A rotor-surface dual-mode wind-resistant attitude holding system configured to perform the control method according to any one of claims 1 to 8, characterized in that, The method comprises: A sensor data acquisition module acquires real-time data of the attitude angle (roll angle, pitch angle, yaw angle), angular rate, position, height, and environmental wind speed and direction of the unmanned aerial vehicle through an IMU (inertial measurement unit), GPS, barometer, and airborne anemometer; A strong wind disturbance judgment module accurately identifies the strong wind disturbance state through a dual judgment mechanism of wind speed threshold and attitude change rate. When the wind speed exceeds the set threshold and the attitude angle change rate is abnormal, it is determined that the strong wind disturbance state is entered; A dynamic target attitude calculation module calculates the "dynamic target balance attitude" in real time based on the current wind field conditions and flight mode; The attitude is not an absolute horizontal zero-degree attitude, but an optimal attitude that can make the unmanned aerial vehicle maintain stability and consume the least energy in the current wind field; An intelligent return-to-center control module adopts a double-layer control architecture of outer loop-inner loop. The outer loop generates expected angular rate instructions based on the dynamic target attitude, and the inner loop realizes rapid angular rate tracking control; A safety monitoring protection module monitors the power system state in real time, automatically degrades control requirements or starts an emergency program when the load is too large, and ensures flight safety; A flight mode adaptive module adopts different control strategies and parameters in the multi-rotor mode and fixed-wing mode according to the characteristics of the compound wing.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to realize the control method of any one of claims 1-8.

Images