A tilt-rotor unmanned aerial vehicle wind-resisting transition control method based on motor current feedback

Abstract

The application discloses a kind of based on motor current feedback's tilting rotor unmanned aerial vehicle wind resistance transition control method, belong to unmanned aerial vehicle flight control technical field.This method is in the transition phase of unmanned aerial vehicle from rotor mode to fixed-wing mode conversion, the current value of two sides tilting motor is collected in real time by electronic speed regulator;Current difference value is calculated and compared with preset wind resistance starting threshold;When exceeding threshold, it is judged to encounter lateral airflow disturbance, and corresponding attitude correction compensation is obtained by table lookup method, directly superimposed into attitude control loop to drive aileron or adjust speed difference.The application directly reuses the current signal of power system, without adding external sensors such as air speed tube, using the leading response characteristics of motor load to airflow, realizes the feedforward compensation to wind disturbance.The method has the advantages of low cost, fast response, advanced perception, simple engineering implementation, significantly improves the attitude stability of tilting transition stage under the disturbance of crosswind, headwind and the like.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) flight control technology, specifically relating to a method for attitude stabilization and wind interference resistance control of a vertical take-off and landing (VTOL) fixed-wing UAV during the transition from rotor flight mode to fixed-wing flight mode. Background Technology

[0002] Tiltrotor drones combine the advantages of multi-rotor vertical takeoff and landing with fixed-wing long endurance. Their key technical challenge lies in the "transition phase," specifically the process of the rotor tilting from vertically upward to horizontally forward. During this phase, the drone's aerodynamic layout and power configuration are in a state of continuous change, resulting in complex and unstable fuselage dynamics. Therefore, they are highly susceptible to asymmetric airflow disturbances such as crosswinds and gusts, which can lead to attitude instability and even flight accidents.

[0003] In existing technologies, mainstream wind resistance control solutions mainly suffer from the following drawbacks: 1. High sensor dependence, low cost and poor reliability: Common solutions (such as the standard VTOL configuration of the open-source flight controller PX4 Autopilot) typically rely on external sensors such as pitot tubes to measure airspeed and indirectly estimate wind field. These sensors not only increase additional hardware costs, weight, and aerodynamic drag, but are also highly susceptible to damage during drone takeoff, landing, transportation, or collisions. If the pitot tube becomes blocked or malfunctions, the flight control system may mistakenly trigger the stall protection program based on incorrect wind speed information, leading to a rapid deterioration of the drone's attitude or even loss of control.

[0004] 2. Control Response Lag: If attitude feedback control relies solely on the inertial measurement unit (IMU, including gyroscopes and accelerometers) within the flight controller, the control system can only begin correction after the UAV's fuselage has been blown by the wind to the point of producing a measurable deviation in attitude angle or angular velocity. This feedback control mode inherently suffers from response lag (typically exceeding 200 milliseconds), making it difficult to effectively suppress sudden gusts of wind interference.

[0005] 3. Complex algorithms and high engineering implementation difficulty: Some academic research uses complex algorithms such as adaptive control, observers, or neural networks to estimate wind fields and perform compensation online. Although these methods are theoretically feasible, they require high computing power from airborne computing resources, involve complex parameter tuning, and make it difficult to guarantee stability, reliability, and maintainability in practical engineering applications.

[0006] Therefore, there is an urgent need in the field for a wind-resistant transition control method that can eliminate the dependence on external wind speed sensors, achieve rapid response, and is easy to engineer. Summary of the Invention

[0007] 1. Technical problem to be solved: The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and provide a wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback. This method eliminates the need for additional wind field sensors, utilizing the current signal from the UAV's own power system to achieve advanced perception and rapid compensation for airflow interference, thereby significantly improving wind resistance and flight safety during the transition phase.

[0008] 2. Technical Solution: To solve the above problems, the present invention adopts the following technical solution.

[0009] A wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback is applied to the transition phase of a UAV switching from rotor mode to fixed-wing mode, and includes the following steps: S1: During the transition phase, the current value of the left tilt motor is collected in real time via the electronic speed controller (ESC). I L and the current value of the right tilt motor I R ; S2: Calculate the current difference Δ between the left and right motors. I =∣ I L - I R |; S3: The current difference Δ I With the preset wind resistance start-up threshold I th Compare; S4: When Δ I > I th At that time, it was determined that the drone was subjected to lateral airflow interference, and based on I L and I R The size relationship determines the wind direction and source; S5: Based on the current difference Δ I The magnitude of the value is determined by looking up a table to obtain the corresponding attitude correction compensation amount, which is then superimposed on the attitude control loop of the UAV to drive the aileron control surfaces or adjust the speed difference between the left and right motors to counteract airflow interference.

[0010] Furthermore, the method also includes an overload protection step: when any motor current value is detected to continuously exceed a safety threshold... I max If the preset time T is exceeded, the control system immediately stops the rotation of the tilting mechanism and forcibly increases the total throttle output to maintain the flight altitude.

[0011] Preferably, before step S2, the acquired raw current data is processed by moving average filtering to filter out high-frequency electromagnetic noise interference.

[0012] Preferably, the lookup table method employs a piecewise linear mapping strategy, based on the current difference Δ I For different intervals, determine the corresponding attitude correction compensation amount. Specifically, this includes: When 0 < Δ I ≤ I th When the corresponding attitude correction compensation is zero, this interval is called the dead zone, which is used to ignore system noise. when I th < Δ I ≤ I sat At that time, the corresponding attitude correction compensation amount C With (Δ) I - I th The relationship is linear, and the compensation amount is... C = K p ×(Δ I - I th ),in K p The preset proportional gain coefficient, I sat This is the saturation threshold; When Δ I > I sat At that time, the corresponding attitude correction compensation amount remains constant at the preset maximum compensation value. C max This is to avoid stalling caused by fully engaging the rudder.

[0013] Furthermore, when the current values ​​of the motors on both the left and right sides are detected... I L and I R If the rise in current exceeds a frontal wind detection threshold within a preset time period, and the current difference Δ I Less than the wind resistance activation threshold I th When the wind is strong enough to be either a pure headwind or a direct wind, the attitude control loop is informed to output a pitch angle compensation command that causes the nose to pitch down.

[0014] Furthermore, based on the current difference ΔI The fluctuation frequency or rate of change is dynamically adjusted to adjust the window length of the moving average filter to adapt to different wind conditions such as steady airflow or turbulent flow.

[0015] Preferably, in step S1, the motor current value is collected in real time by an electronic speed controller that supports digital communication protocols such as DSHOT and CAN, so as to achieve high bandwidth and low latency data transmission.

[0016] The method can be extended to unmanned aircraft configurations with multiple symmetrically distributed tilt rotors (such as quadcopter tilt rotors). In this case, when calculating the current difference in step S2, the total current of the left group of motors and the total current of the right group of motors should be calculated separately, and the absolute value of their difference should be obtained.

[0017] 3. Beneficial effects: Compared with the prior art, the technical solution provided by this invention has the following advantages: (1) Significantly reduce cost and system complexity: This invention directly reuses the inherent current feedback data of the electronic speed controller in the UAV power system, without the need to install any additional hardware such as airspeed tubes and angle of attack sensors, which fundamentally reduces the system manufacturing cost, overall weight and potential failure points, and achieves lightweight and high reliability of the system.

[0018] (2) Achieving advanced sensing and rapid feedforward compensation: The motor operating current is extremely sensitive to changes in the propeller aerodynamic load. When crosswind interference occurs, the difference in load between the left and right motors will immediately be reflected in the current difference. This process is much faster than the process of attitude change of the fuselage due to force (it can lead by about 100-200 milliseconds). This invention uses this current difference signal for discrimination and compensation, changing the traditional "hysteresis feedback control" to "advanced feedforward control", which greatly shortens the system response time.

[0019] (3) The control logic is simple and reliable and easy to integrate into engineering: The core control logic is based on threshold judgment and table lookup compensation method, which has a small amount of computation and low requirements for processor performance. It does not require complex online identification or iterative calculation, making it very suitable for stable and real-time operation on low-cost microcontrollers such as STM32, which greatly improves the practicality and industrialization feasibility of the method.

[0020] (4) Strong robustness and adaptability to various complex wind conditions: The motor noise itself is effectively filtered out by setting a reasonable dead zone; the piecewise linear mapping and saturation limiting strategy take into account both the smooth compensation of small disturbances and the safe limit of large disturbances; and it can intelligently distinguish different disturbance modes such as crosswind and headwind, and execute targeted compensation strategies, which significantly improves the attitude stability and flight safety in complex wind field environment during the transition phase.

[0021] It should be noted that the structures not described in this invention are not related to the design points and improvement directions of this invention, and are the same as or can be implemented using existing technologies, so they will not be elaborated here. Attached Figure Description

[0022] Figure 1 The logic flowchart of the control method provided in the embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram illustrating the physical response principle and current waveform simulation under crosswind interference in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the hardware connection and signal flow of the system of the present invention.

[0025] Figure 4 The current difference Δ in this invention I Piecewise linear mapping curve of attitude correction compensation amount C. Detailed Implementation

[0026] The technical solution of the present invention will now be described in detail and completely with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can fully understand and implement the present invention. It should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0027] Example 1: Basic Control Flow under Crosswind Interference like Figure 1 and Figure 3 As shown, this embodiment uses a typical dual-rotor tilt-rotor UAV as an example. Its system includes a power battery, a flight control main processing unit, a left electronic speed controller (ESC), a right electronic speed controller (ESC), a left tilt motor, a right tilt motor, and corresponding servos (not labeled in the figure). The flight control main processing unit establishes bidirectional communication with the left and right ESCs via a digital bus (such as DSHOT). While driving the motors (Motor_L, Motor_R), the ESCs can transmit the motor bus current data back to the flight control unit in real time.

[0028] 1. Data Acquisition and Preprocessing: When the UAV enters transition mode (rotor tilt angle varies between 0° and 90°), the flight controller reads the current of the left motor at a frequency of 50Hz. I L and the current of the motor on the right I R To suppress the high-frequency electromagnetic noise introduced by the PWM drive, the original current data is filtered by a moving average with a window length of 5.

[0029] It should be noted that real-time acquisition of accurate motor current values ​​is fundamental to the realization of this invention. Preferably, this is achieved through an electronic speed controller (ESC) supporting bidirectional digital communication protocols such as DSHOT and CAN, which provides a high-bandwidth, low-latency data channel. Those skilled in the art can also use other speed control modules with current feedback functionality or external high-precision current sensors to achieve the same purpose.

[0030] 2. Threshold setting and judgment: Wind resistance activation threshold I th The setting is based on the current fluctuation range of the motor under steady-state operation (such as hovering). For example, if the measured steady-state current noise peak is approximately ±0.3A, then the setting is... I th =0.5A, forming an effective signal dead zone. The system calculates the filtered current difference Δ in real time. I =∣ I L - I R |

[0031] 3. Wind field identification and table lookup compensation: Suppose that during the transition, the drone is suddenly struck by a crosswind of approximately 6 m / s on its right side. Figure 2 As shown, the right propeller becomes the windward side, and the increased effective incoming flow velocity leads to increased aerodynamic drag and a higher motor load. I R Consequently, the load increases; the left propeller is on the leeward side, so the load is relatively reduced. I L It may decrease slightly or remain unchanged. The system detected... I R > I L And Δ is calculated I =1.8A.

[0032] Due to Δ I (1.8A)> I th (0.5A), the system determined that it encountered crosswind interference from the right.

[0033] In practical engineering implementation, the above piecewise linear mapping relationship can be calculated in real time in the controller using the formula, or a set of discrete (Δ) values ​​can be pre-calculated and stored based on this relationship. I C) Numerical lookup table. The "table lookup method" described in this invention covers both of these implementation methods.

[0034] 4. Feedforward compensation execution: The system performs feedforward compensation based on pre-stored piecewise linear mapping relationships (such as...). Figure 4 (As shown) Determine the compensation amount. For Δ I =1.8A, which falls within the linear compensation region (0.5A < Δ). I ≤ 4.0A). If the proportional gain coefficient K p If the calibration is 1.2° / A, then the calculated roll angle compensation is C = 1.2 × (1.8 - 0.5) ≈ 1.56°. This compensation command is in the "left roll" direction (i.e., commanding the right aileron to deflect upwards and the left aileron to deflect downwards).

[0035] 5. Control Effect: This compensation is immediately fed forward and superimposed onto the flight controller's attitude control output. Therefore, before the IMU detects the rightward roll deviation of the fuselage due to right-side winds, the flight controller has already issued control surface commands to generate a leftward roll recovery torque to counteract the right roll torque. Based on principle analysis and simulation results, this method can significantly suppress the maximum fuselage roll deviation under these wind conditions from approximately 4.2° under traditional PID feedback control to approximately 1.2°, and enable the attitude to stabilize in a shorter time.

[0036] Example 2: Multi-wind condition handling and adaptive protection mechanism This embodiment further illustrates the adaptability of the present invention to complex wind field environments.

[0037] 1. Handling interference from pure headwinds or direct winds: When a drone encounters a strong headwind head-on, the load on the left and right propellers will increase symmetrically, leading to... I L and I R Simultaneous and significant increase, while at this time Δ I The value may be very small (less than) I th The system identifies headwinds by monitoring the rate and magnitude of current rise. For example, a valid 'headwind detection threshold' can be set when the current values ​​of both the left and right motors rise by more than 20% of the steady-state current value under that throttle condition within 0.1 seconds. Once a strong headwind is detected, the system does not perform roll compensation but instead outputs an appropriate "pitch down" command to the attitude control loop to reduce the angle of attack and help the drone maintain or increase its airspeed to more effectively penetrate the headwind.

[0038] 2. Turbulence (turbulent flow) disturbance handling: In turbulent flow, the direction and velocity of the airflow change rapidly and randomly, leading to Δ I The signal exhibits high-frequency, large-amplitude oscillations. To prevent the servo motor from following these oscillations with high-frequency jitter, increasing energy consumption and potentially causing resonance, the control system has an adaptive filtering function. For example, when Δ is detected... IWhen the rate of change or number of zero crossings exceeds a certain threshold within 1 second, the window length of the moving average filter is automatically increased from the default 5 points to 10 points or higher, thereby effectively smoothing control commands and enhancing the system's robustness in non-steady airflow.

[0039] 3. Motor overload protection mechanism: To prevent the motor from overheating or being damaged due to continuous high current operation under extreme wind conditions, the system sets a safe current threshold. I max =30A and continuous monitoring time T =1 s. If the current value of any motor continues to exceed 30A for 1 second, the control system will immediately pause the operation of the tilt servo mechanism, lock the current rotor tilt angle, and at the same time slightly increase the total throttle output to compensate for the lift loss that may be caused by the suspension of tilting, prioritizing the safety of flight altitude, and resume normal control procedures after the current drops back to a safe range.

[0040] Example 3: Extended application of the method in different configurations The principles of this invention are not limited to the twin-rotor tilt configuration and can be easily extended to other vertical takeoff and landing (VTOL) configurations.

[0041] For a quadcopter tilt-rotor drone: it has four tilt motors: left front, left rear, right front, and right rear. The sum of the currents from the left front and left rear motors can be used as the "total current on the left side". I L_total The sum of the currents of the front right and rear right motors is taken as the "total current on the right side". I R_total The subsequent current difference calculation, threshold judgment, and compensation logic are completely consistent with those in Example 1, i.e., calculating Δ... I =∣ I L_total - I R_total |

[0042] For tail-sitter drones, their wind resistance also relies on the thrust / load difference between the left and right propulsion systems. Therefore, by collecting and comparing the operating currents of the left propulsion motor (or motor assembly) and the right propulsion motor (or motor assembly), the method described in this invention can also be applied to achieve wind resistance compensation.

[0043] Expected Experimental Results To verify the principle effectiveness and expected advantages of the control method of this invention, a systematic simulation analysis was conducted based on a high-fidelity dynamics model of an unmanned aerial vehicle (UAV) and wind tunnel environment simulation. Table 1 below shows the expected performance comparison data between the method of this invention and the traditional IMU feedback control method under set crosswind interference. These data aim to reveal the significant improvements brought about by this invention in principle.

[0044] Table 1: Comparison of Expected Wind Resistance Effects Based on Simulation Analysis

[0045] Comparison of quantification effects: Wind resistance response time: The response time of the traditional IMU feedback scheme is about 250ms. This invention uses the current lead signal to shorten the response time to about 60ms.

[0046] Attitude stabilization time: The time required to recover from encountering gust interference to regain horizontally stable flight is reduced from approximately 1.5 seconds in the traditional scheme to approximately 0.4 seconds.

[0047] The above expected data fully demonstrate that the method of the present invention has outstanding effects in improving wind resistance, control response speed and flight safety during the tilt transition phase.

[0048] Finally, it should be noted that the above embodiments are only used to fully illustrate the technical solution and core idea of ​​the present invention, and are not intended to limit the scope of protection of the present invention. For those skilled in the art, several improvements, equivalent substitutions, or modifications can be made without departing from the principle of the present invention, and these improvements, equivalent substitutions, or modifications should also be considered within the scope of protection of the present invention. For example, adaptive adjustments to the specific form of the filtering algorithm, the specific values ​​of the mapping table, and the distribution ratio of the compensation amount between the aileron and the differential speed all fall within the scope defined by the claims of the present invention.

Claims

1. A wind-resistant transition control method for tilt-rotor unmanned aerial vehicles based on motor current feedback, characterized in that, Includes the following steps: S1: During the transition from rotorcraft mode to fixed-wing mode of the UAV, the current value of the left tilt motor is collected in real time through the electronic speed controller. I L and the current value of the right tilt motor I R ; S2: Calculate the current difference Δ between the left and right motors. I =∣ I L - I R |; S3: The current difference Δ I With the preset wind resistance start-up threshold I th Compare; S4: When Δ I > I th At that time, it was determined that the drone was subjected to lateral airflow interference, and based on I L and I R The relative sizes of the winds indicate their direction and origin. S5: Based on the current difference Δ I The magnitude of the value is determined by looking up a table to obtain the corresponding attitude correction compensation amount, which is then superimposed on the attitude control loop of the UAV to drive the aileron control surfaces or adjust the speed difference between the left and right motors to counteract airflow interference.

2. The wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 1, characterized in that, It also includes overload protection: when any motor current value is detected to continuously exceed the safety threshold. I max If the preset time T is exceeded, the control system immediately stops the rotation of the tilting mechanism and forcibly increases the total throttle output to maintain the flight altitude.

3. The wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 1, characterized in that, Before step S2, the collected raw current data is subjected to moving average filtering.

4. The wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 1, characterized in that, The table lookup method employs a piecewise linear mapping strategy, based on the current difference Δ. I For different intervals, determine the corresponding attitude correction compensation amount.

5. A wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 4, characterized in that, The table lookup method employs a piecewise linear mapping strategy: When 0 < Δ I ≤ I th When the corresponding attitude correction compensation is zero, this interval is a dead zone; when I th < Δ I ≤ I sat At that time, the corresponding attitude correction compensation amount C With (Δ) I - I th The relationship is linear, and the compensation amount is... C = K p ×(Δ I - I th ),in K p The preset proportional gain coefficient, I sat The saturation threshold; When Δ I > I sat At that time, the corresponding attitude correction compensation amount remains constant at the preset maximum compensation value. C max .

6. The wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 1, characterized in that, When the current values ​​of the motors on both the left and right sides are detected I L and I R If the rise in current exceeds a frontal wind detection threshold within a preset time period, and the current difference Δ I Less than the wind resistance activation threshold I th When encountering a pure headwind or a direct wind, the system determines that it is encountering a pitch angle compensation command and outputs it to the attitude control loop.

7. A wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 3, characterized in that, According to the current difference Δ I The fluctuation frequency is dynamically adjusted to change the window length of the moving average filter.

8. A wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 1, characterized in that, In step S1, the motor current value is collected in real time by an electronic speed controller that supports digital communication protocols.

9. A wind-resistant transition control method for tilt-rotor UAVs based on motor current feedback according to claim 1, characterized in that, The method is applicable to UAVs with multiple symmetrically distributed tilt rotors. In step S2, when calculating the current difference, the total current of the left group of motors and the total current of the right group of motors are calculated respectively, and the absolute value of their difference is obtained.

Images