A method for wind-resistant stability control of unmanned aerial vehicles

Abstract

This invention relates to the field of adaptive control technology, specifically to a method for wind-resistant stability control of unmanned aerial vehicles (UAVs). The method includes the following steps: extracting the absolute parameter of the yaw angle error integral, calculating the limiting yaw boundary by combining roll and pitch moments to replace the original integral, outputting a restricted yaw command, adding a symbol to the hovering lift to establish a thrust command, and taking the duty cycle to send to the speed controller to drive the motor for attitude stabilization. In this invention, by combining the UAV's roll and pitch channel torque data to calculate the equivalent horizontal thrust consumption, and by back-calculating the limiting yaw moment boundary value based on the maximum thrust of a single motor, the underlying physical limits are mapped upwards, and the accumulated yaw error integral is truncated for overlay update. The target thrust command is reconstructed by fusing the UAV's hovering lift, matched with the duty cycle parameters, and sent to the drive terminal. This completely eliminates the conflict between invalid accumulation of yaw integrals and motor speed saturation under multi-axis coupling, ensuring the UAV's wind-resistant attitude balance and stability in complex wind conditions.

Description

Technical Field

[0001] This invention relates to the field of adaptive control technology, and in particular to a method for wind-resistant stability control of unmanned aerial vehicles (UAVs). Background Technology

[0002] Adaptive control technology refers to a class of control methods that maintain stable system operation by identifying system parameters online and adjusting the control law in real time based on the identification results when the structural parameters of the controlled object or external environmental conditions change. Among these methods, the UAV wind disturbance stability control method addresses the problem of attitude angle deviation and trajectory deviation caused by gusts, continuous crosswinds, or vertical airflows during UAV flight. It obtains roll, pitch, yaw angle, and angular velocity data through an onboard inertial measurement unit, and combines this with altitude and position information obtained from a barometer and satellite positioning module. First, a rigid body dynamics equation including wind disturbance terms is established. Then, an error variable is constructed based on the deviation between the desired and actual attitude angles. A disturbance estimation term is introduced into the controller, and a compensation quantity is generated by integrating the angular velocity deviation. This compensation quantity is then superimposed on the speed distribution of each motor to change the lift and torque distribution generated by each rotor. The process of calculating the attitude error, updating the compensation quantity, and outputting motor drive signals is repeated according to a fixed control cycle to adjust the flight attitude under wind disturbance conditions.

[0003] In the current process of wind-resistant stability control of UAVs, when dealing with continuous wind-resistant operation modes, error variables are directly constructed based on attitude deviations and disturbance estimation terms are introduced. The compensation amount is directly generated by integrating the angular velocity deviation and blindly added to the motor speed distribution. Since the multi-attitude axial control shares the rotor motor and roll and pitch occupy a large amount of output capacity, the system lacks a mechanism for overall constraint on the maximum thrust boundary of the underlying motor. This causes the yaw error integral to accumulate blindly. After superposition, the target speed is very likely to exceed the hardware limit, causing deep saturation conflict. Ultimately, this leads to the failure of wind-resistant lift distribution and flight attitude instability, deviating from the intended flight path. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for wind-resistant stability control of unmanned aerial vehicles (UAVs).

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for wind-resistant stability control of unmanned aerial vehicles (UAVs), comprising the following steps:

[0006] S1: Monitor the historical cumulative yaw angle error integral parameters of the UAV airborne flight control processor's anti-wind disturbance yaw closed loop, perform absolute value extraction on the yaw angle error integral parameters, remove the original positive and negative signs and retain the absolute size to generate yaw wind resistance integral absolute parameters;

[0007] S2: Collect the current wind-resistant roll channel torque and wind-resistant pitch channel torque of the UAV, calculate the equivalent horizontal thrust consumption and the boundary value of the ultimate yaw moment, select the boundary value of the ultimate yaw moment to replace the yaw wind-resistant integral absolute parameter, and output the constrained yaw moment constraint command.

[0008] S3: Extract the original sign of the yaw angle error integral parameter and append it to the restricted yaw moment constraint command write-back overwrite storage address, collect the UAV hovering lift and replace the original single motor maximum thrust, and establish the target wind-resistant thrust drive command.

[0009] S4: Collect the UAV preset thrust duty cycle data table, perform a search action according to the target wind-resistant thrust drive command, extract the corresponding pulse width modulation duty cycle parameters and send them to the rotor motor electronic speed controller, and obtain the terminal motor hardware drive command.

[0010] S5: Based on the terminal motor hardware drive command, combined with the response drive signal of the wind-resistant rotor motor at the end of the UAV arm and the electronic speed controller, the rotor mechanism of the wind-resistant rotor motor is triggered to rotate, and the attitude balance state of the UAV is adjusted and the trajectory is corrected, and the UAV wind disturbance stability control record is output.

[0011] As a further aspect of the present invention, the calculation of equivalent horizontal thrust consumption specifically refers to extracting the absolute values ​​of the anti-wind roll channel torque and the anti-wind pitch channel torque of the UAV, adding them together, and dividing by the length of the UAV arm.

[0012] As a further aspect of the present invention, the yaw wind-resistant integral absolute parameters include the heading drift magnitude, attitude offset magnitude, and cumulative error modulus; the constrained yaw moment constraint commands include output truncation threshold, dynamic compensation limit, and safety correction quota; the target wind-resistant thrust drive commands include desired speed reference, expected power distribution, and lift adjustment ratio; the terminal motor hardware drive commands include waveform trigger pulse, current control level, and frequency adjustment sequence; and the UAV wind disturbance stability control record includes attitude balance state scalar, trajectory correction feedback quantity, and hovering stability variables.

[0013] As a further aspect of the present invention, the specific steps for obtaining the yaw wind resistance integral absolute parameter are as follows:

[0014] S111: Monitor the operation data stream of the anti-wind disturbance yaw closed loop inside the UAV's airborne flight control processor, extract the historical cumulative yaw angle error integral parameter value sequence, sort and verify the integral values ​​of each sampling period according to the timestamp order, and generate the yaw error integral sequence.

[0015] S112: Based on the yaw error integral sequence, perform a sign bit splitting operation on each integral value, convert the positive and negative signs into independent sign identifier fields and retain the corresponding numerical fields, establish a mapping between the sign identifier fields and the numerical fields, and generate an integral sign numerical mapping set.

[0016] S113: Based on the numerical mapping set of integral symbols, perform absolute value operation on the numerical fields, convert the signed integral value into an unsigned numerical value and retain the symbol identification field in an independent storage area, keep the correspondence of each time node unchanged and complete the data structure reorganization to generate the yaw wind resistance integral absolute parameter.

[0017] As a further aspect of the present invention, the specific steps for obtaining the restricted yaw moment constraint command are as follows:

[0018] S211: Collect the wind-resistant roll channel torque and wind-resistant pitch channel torque of the UAV, perform absolute value calculation on the torque of the two channels respectively, perform addition operation on the two absolute values ​​to form the total horizontal resultant torque, extract the UAV arm length and perform division operation on the total horizontal resultant torque to generate the equivalent horizontal thrust consumption value;

[0019] S212: Based on the equivalent horizontal thrust consumption value, obtain the maximum thrust of a single motor, the reverse torque conversion parameter, the motor efficiency coefficient, the current motor speed, the rated speed reference value, and the air density reference value, and calculate the boundary value of the limiting yaw moment.

[0020] S213: Based on the boundary value of the limiting yaw moment, perform a numerical replacement operation on the absolute integral parameter of the yaw wind resistance, write the boundary value of the limiting yaw moment into the moment constraint field of the yaw control channel and cover the original integral parameter input channel, keep the field length of the control command data structure consistent and complete the command reconstruction to obtain the restricted yaw moment constraint command.

[0021] As a further aspect of the present invention, the formula for calculating the boundary value of the limiting yaw moment is as follows:

[0022] ;

[0023] in, This represents the boundary value of the limiting yaw moment. This indicates the maximum thrust of a single motor. This represents the equivalent horizontal thrust consumption value. Indicates the inverse torque conversion parameter. This represents the motor efficiency coefficient. Indicates the current motor speed. This indicates the reference value for the rated speed.

[0024] As a further aspect of the present invention, the specific steps for obtaining the target wind-resistant thrust driving command are as follows:

[0025] S311: Extract the original sign bit identifier of the yaw angle error integral parameter, split the sign bit and the value bit and generate a sign bit identifier sequence, detect the structure of the restricted yaw moment constraint command data frame and locate the write-back overwrite storage address field, append the sign bit identifier sequence to the restricted yaw moment constraint command torque value field and perform a sign composite operation to generate a sign composite yaw moment value.

[0026] S312: Based on the symbolic composite yaw moment value, read the restricted yaw moment constraint instruction storage address mapping table, perform a fixed-point write operation on the write-back overwrite storage address field and verify the consistency of the address segment check code, complete the data frame overwrite replacement and keep the field byte length unchanged, and generate a write-back yaw moment control frame.

[0027] S313: Collect the hovering lift value of the UAV, detect the maximum thrust field of a single motor and perform a value substitution operation, write the hovering lift value into the original maximum thrust field of a single motor and establish an association mapping relationship between the thrust value and the written-back yaw torque control frame, reconstruct the drive output based on the ratio of the thrust value and the torque value, and generate the target wind-resistant thrust drive command.

[0028] As a further aspect of the present invention, the specific steps for obtaining the terminal motor hardware drive instruction are as follows:

[0029] S411: Collect the UAV preset thrust duty cycle data table, extract the pulse width modulation duty cycle parameters that match the target wind-resistant thrust drive command, perform a search operation based on the thrust value in the target wind-resistant thrust drive command, obtain the corresponding pulse width modulation duty cycle value, and generate a pulse width modulation duty cycle parameter sequence.

[0030] S412: Based on the pulse width modulation duty cycle parameter sequence, read the communication interface of the rotor motor electronic speed controller, and send it to the rotor motor electronic speed controller one by one according to the matching duty cycle value to generate motor speed control command;

[0031] S413: Receive the status confirmation signal from the rotor motor electronic speed controller to confirm that the motor speed control command has been successfully transmitted and executed, and generate the terminal motor hardware driver command based on the motor feedback data.

[0032] As a further aspect of the present invention, the specific steps for obtaining the UAV wind disturbance resistance stability control record are as follows:

[0033] S511: Based on the terminal motor hardware drive command, monitor the response signal of the wind-resistant rotor motor at the end of the UAV arm, collect the rotation status data of the rotor mechanism of the rotor motor, trigger the rotation action of the motor rotor mechanism, start the UAV attitude balance adjustment module, and generate a preliminary attitude adjustment signal.

[0034] S512: Based on the preliminary attitude adjustment signal, the attitude of the UAV is corrected in real time, and the attitude change is balanced and adjusted by combining the rotor motor speed and thrust value, and a dynamic trajectory correction command is generated.

[0035] S513: Combining the trajectory correction command and the initial attitude adjustment signal, record the dynamic adjustment data and stable state parameters during the control process and output the control log to generate a UAV anti-wind disturbance stability control record.

[0036] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0037] In this invention, the equivalent horizontal thrust consumption is calculated by combining the roll and pitch channel torque data of the UAV. The boundary value of the limit yaw torque is derived from the maximum thrust of a single motor. The underlying physical limit is mapped upward and the cumulative yaw error integral is truncated for coverage update. The target thrust command for UAV hovering lift reconstruction is matched with the duty cycle parameters and sent to the drive terminal. A hardware reverse calculation and integral truncation collaborative processing flow is constructed to completely eliminate the conflict between invalid accumulation of yaw integral and motor speed saturation under multi-axis coupling. The torque distribution space is accurately coordinated, which fundamentally ensures the overall wind resistance attitude balance and stability of the UAV in complex wind field environments. Attached Figure Description

[0038] Figure 1 This is a flowchart of the main steps of the present invention;

[0039] Figure 2 This is a flowchart of the process for obtaining the integral absolute parameter of yaw wind resistance in this invention;

[0040] Figure 3 This is a flowchart of the process for obtaining the constrained yaw moment command of the present invention;

[0041] Figure 4 This is a flowchart illustrating the process of obtaining the wind-resistant thrust drive command for the present invention.

[0042] Figure 5 This is a flowchart of the process for obtaining the terminal motor hardware driver instruction in this invention;

[0043] Figure 6 This is a flowchart of the process for obtaining the wind disturbance resistance stability control record of the UAV according to the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0045] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0046] Please see Figure 1 A method for wind-resistant stability control of unmanned aerial vehicles (UAVs) includes the following steps:

[0047] S1: Monitor the historical cumulative yaw angle error integral parameters of the UAV airborne flight control processor's anti-wind disturbance yaw closed loop, perform absolute value extraction on the yaw angle error integral parameters, remove the original positive and negative signs and retain the absolute size to generate yaw wind resistance integral absolute parameters;

[0048] S2: Collect the current wind-resistant roll channel torque and wind-resistant pitch channel torque of the UAV, extract the absolute values ​​of the two, add them together and divide by the length of the UAV arm to calculate the equivalent horizontal thrust consumption, subtract the equivalent horizontal thrust consumption from the maximum thrust of a single motor and multiply by the inverse torque conversion parameter to obtain the limit yaw moment boundary value, select the limit yaw moment boundary value to replace the yaw wind-resistant integral absolute parameter, and output the constrained yaw moment constraint command;

[0049] S3: Extract the original sign of the yaw angle error integral parameter and append it to the restricted yaw moment constraint command to write back and overwrite the storage address, collect the hovering lift of the UAV and replace the original single motor maximum thrust, and establish the target wind-resistant thrust drive command.

[0050] S4: Collect the UAV preset thrust duty cycle data table, perform a search action according to the target wind-resistant thrust drive command, extract the corresponding pulse width modulation duty cycle parameters and send them to the rotor motor electronic speed controller to obtain the terminal motor hardware drive command.

[0051] S5: Based on the terminal motor hardware drive command, combined with the response drive signal of the wind-resistant rotor motor at the end of the UAV arm and the electronic speed controller, the rotor mechanism of the wind-resistant rotor motor is triggered to rotate, so as to adjust the attitude balance state and correct the trajectory of the UAV, and output the UAV wind disturbance stability control record.

[0052] The integral absolute parameters of yaw wind resistance include the magnitude of heading drift, the magnitude of attitude deviation, and the magnitude of cumulative error. The constrained yaw moment control commands include the output cutoff threshold, dynamic compensation limit, and safety correction quota. The target wind resistance thrust drive commands include the desired speed reference, the expected power distribution, and the lift adjustment ratio. The terminal motor hardware drive commands include waveform trigger pulses, current control levels, and frequency adjustment sequences. The UAV wind disturbance stability control records include attitude balance state scalars, trajectory correction feedback quantities, and hovering stability variables.

[0053] Please see Figure 2 The specific steps of S1 are as follows:

[0054] S111: Monitor the operation data stream of the anti-wind disturbance yaw closed loop inside the UAV's airborne flight control processor, extract the historical cumulative yaw angle error integral parameter value sequence, sort and verify the integral values ​​of each sampling period according to the timestamp order, and generate the yaw error integral sequence.

[0055] A serial peripheral interface bus communication link is established with the UAV's onboard flight control processor. The serial clock frequency is configured to 10 MHz, and the data width to be 8 bits. A direct memory access controller is used to continuously read the operating data stream of the internal wind-damping yaw closed-loop circuit at a fixed trigger frequency of 100 Hz. A 4096-byte circular buffer is allocated in the internal static random access memory. The extracted data stream undergoes frame header matching and checksum verification. The communication protocol encapsulation is stripped, and the 32-bit floating-point historical cumulative yaw angle error integral parameter values ​​at offset addresses 0x1A to 0x1D are located and extracted. A doubly linked list data structure containing timestamp parameters and error integral parameters is established. The numerical sequence of 500 consecutively acquired sampling periods is sorted according to timestamp order using a quicksort algorithm. The system sets up array head and tail pointers, selects the timestamp corresponding to the middle physical address as the benchmark comparison value, traverses the left pointer to the right to find timestamps greater than the benchmark value, and the right pointer traverses the left to find timestamps less than the benchmark value. Once both sides lock the target, the memory address data is swapped. This process is recursively divided into left and right subsequences and processed in the same way until the subsequence length is reduced to 1, thus completing the absolute temporal alignment of the physical storage space. The sorted doubly linked list is traversed, and the timestamp difference between adjacent nodes is calculated to perform continuity verification. The continuity verification threshold is set to 15 milliseconds, based on the statistical distribution of the UAV's yaw axis rotational inertia and the physical system response delay at a 50 Hz Nyquist control frequency. When a timestamp difference greater than 15 milliseconds is detected, a data packet loss or disconnection is determined. At this point, two valid sampling points before and after the breakpoint are immediately extracted and substituted into a Lagrange cubic polynomial interpolation algorithm. Weighted polynomial summation is used to calculate and complete the integral values ​​corresponding to the missing timestamp nodes, generating a yaw error integral sequence with absolutely equidistant continuity on the time axis.

[0056] S112: Based on the yaw error integral sequence, perform sign bit splitting operation on each integral value, convert positive and negative signs into independent sign identifier fields and retain the corresponding numerical fields, establish a mapping between sign identifier fields and numerical fields, and generate an integral sign numerical mapping set.

[0057] For the generated yaw error integral sequence, 32-bit single-precision floating-point integral values ​​are loaded one by one into the arithmetic logic unit (ALU) of the central processing unit (CPU) according to memory addresses. A sign bit splitting operation is performed using bitwise operations, calling the hexadecimal mask parameter 0x80000000 and performing a bitwise logical AND operation with the loaded integral values. The result of the logical AND operation is then shifted 31 bits to the right to precisely separate the most significant bit at the 31st bit, i.e., the positive or negative sign bit. When the extracted bit value is 0, it is converted into a positive independent identifier; when the extracted bit value is 1, it is converted into a negative independent identifier. A structure array is instantiated in the dynamic memory area, containing an 8-bit unsigned integer sign identifier field and a 32-bit floating-point numeric field. The separated positive and negative sign identifiers are written to the sign identifier field, and the lower 31 significant data bits of the original integral value are written intact to the corresponding numeric field. A one-to-one strong mapping relationship between the sign identifier field and the numeric field in physical memory is established using array index keys, generating an integral sign-numeric mapping set with discrete storage characteristics.

[0058] S113: Based on the numerical mapping set of integral symbols, perform absolute value operation on the numerical fields, convert the signed integral value into an unsigned numerical value and retain the symbol identification field in an independent storage area, keep the correspondence of each time node unchanged and complete the data structure reorganization to generate the yaw wind resistance integral absolute parameter.

[0059] The system reads the integral sign value mapping set, loads the numerical fields in the structure into the floating-point unit, and performs a bitwise logical AND operation with the numerical field using the hexadecimal mask parameter 0x7FFFFFFF. This absolute value operation forces the 31st bit to be cleared, truncates any negative expression mechanism, and converts all signed integral values ​​into pure unsigned values ​​at the binary level. Subsequently, a memory block release instruction is executed in the storage space, directly erasing the 8-bit unsigned integer sign identifier field in the structure and releasing the bus width resources it occupies. The memory offset address of the current processing node is calculated, and the converted 32-bit unsigned values ​​are rewritten continuously into consecutive read-only data segment memory addresses according to the timestamp sequence in the original mapping set. This strictly maintains the physical timing relationship of the data at each time node, eliminates memory fragmentation, and completes a compact reorganization of the data structure, ultimately generating a yaw wind-resistant integral absolute parameter that only contains positive accumulation characteristics.

[0060] Please see Figure 3The specific steps of S2 are as follows:

[0061] S211: Collect the wind-resistant roll channel torque and wind-resistant pitch channel torque of the UAV, perform absolute value calculation on the torque of the two channels respectively, perform addition operation on the two absolute values ​​to form the total horizontal resultant torque, extract the UAV arm length and perform division operation on the total horizontal resultant torque to generate the equivalent horizontal thrust consumption value;

[0062] The data channel of the three-axis torque sensor of the microelectromechanical system (MEMS) configured at the base of the UAV arm is connected. The system continuously collects the roll torque (resisting lateral crosswind disturbances) and the pitch torque (resisting longitudinal airflow disturbances) of the UAV at a sampling rate of 1000 Hz via the internal integrated circuit bus. A second-order Butterworth low-pass filter with a cutoff frequency of 30 Hz is configured in the microprocessor for each of the two data streams to filter out high-frequency harmonic noise caused by high-frequency rotor vibrations. The filtered roll and pitch torque values ​​are then subjected to low-level binary mask absolute value operations to remove the direction vector characteristics of the torques, retaining only the physical scalar values ​​reflecting the disturbance intensity. The addition register in the arithmetic logic unit is invoked to perform a single-precision floating-point addition operation on the absolute values ​​of the roll and pitch parameters, merging them to generate the total horizontal torque consumed by the UAV to maintain its horizontal attitude. A read command is sent to the electrically erasable programmable read-only memory (EROM) to extract the pre-programmed physical parameters of the UAV arm length. The arm length parameter is obtained from the precise spatial distance measurement results of the vertical axis from the rotor shaft center to the fuselage center of gravity by a three-dimensional coordinate measuring machine during the factory calibration phase. The total horizontal resultant torque value is placed in the dividend register of the divider, and the arm length value is placed in the divisor register. The hardware division instruction is executed to generate the equivalent horizontal thrust consumption value that characterizes the additional horizontal thrust consumed by the UAV to resist horizontal wind disturbance.

[0063] S212: Based on the equivalent horizontal thrust consumption value, obtain the maximum thrust of a single motor, the reverse torque conversion parameters, the motor efficiency coefficient, the current motor speed, the rated speed reference value, and the air density reference value, using the following formula:

[0064] ;

[0065] The boundary values ​​of the limiting yaw moment are obtained through calculation; where... This represents the boundary value of the limiting yaw moment. This indicates the maximum thrust of a single motor. This represents the equivalent horizontal thrust consumption value. Indicates the inverse torque conversion parameter. This represents the motor efficiency coefficient. Indicates the current motor speed. Indicates the reference value of the rated speed;

[0066] The factory calibration firmware section of the UAV rotor propulsion system was consulted to obtain the maximum thrust parameter of a single motor. The dataset from the load sensor on the anti-torque test bench was connected via the communication interface to extract the anti-torque conversion parameter. The calibration process for this anti-torque conversion parameter is as follows: the motor is fixed to the torque test bench, and the ratio of yaw axis reverse torque to vertical thrust under different duty cycles is recorded. A linear regression fitting is performed on multiple sets of discrete ratio data using the least squares method, and the slope of the fitted line is extracted as the parameter setpoint. The input electrical power is calculated by reading the bus current data and phase voltage data from the Hall current sensor built into the electronic speed controller. The ratio is then compared with the mechanical power measured by the dynamometer to obtain the motor efficiency coefficient. A two-way digital communication telemetry link with the electronic speed controller is established, and the revolutions per minute (RPM) segment in the telemetry data frame is parsed to obtain the current motor speed. The rated speed reference value that maintains the optimal lift-to-drag ratio for the entire aircraft, obtained statistically from long-term high-altitude cruise experiments, is read from the flight control parameter table. The digital barometer is initialized to read the atmospheric static pressure and temperature values. Substituting the obtained parameters—maximum thrust of a single motor (25.0), equivalent horizontal thrust consumption (8.5), inverse torque conversion parameter (0.06), motor efficiency coefficient (0.85), current motor speed (4200), and rated speed reference value (5000)—into the above formula, the final result is calculated: the limit yaw torque boundary value is 0.707. The innovation of this formula lies in the introduction of a deviation coefficient between the actual motor speed and the rated speed, dynamically reducing the upper limit of the limit torque output. This allows for sufficient control margin when the motor is operating outside its rated condition, preventing speed saturation and motor stall caused by full-load output. This enhances the rotor attitude stability under extreme wind conditions in the overall system's wind resistance control. The data parameters for the above process are shown in Table 1 below.

[0067] Table 1 Boundary Calculation Parameters for Ultimate Yaw Moment

[0068]

[0069] Table 1 lists the specific values ​​of each parameter in the embodiment and the final calculated limit boundary values. This result shows that, while ensuring horizontal wind resistance, the maximum torque that the current UAV's power distribution system can safely output via the yaw axis is strictly limited to a safe threshold of 0.707, avoiding encroachment on the thrust resources of the roll and pitch channels. This numerical result is then compared with the memory address space where the yaw wind resistance integral absolute parameter is located, preparing for the next instruction reconfiguration step.

[0070] S213: Based on the boundary value of the limiting yaw moment, perform a numerical replacement operation on the integral absolute parameter of yaw wind resistance, write the boundary value of the limiting yaw moment into the moment constraint field of the yaw control channel and overwrite the original integral parameter input channel, keep the field length of the control command data structure consistent and complete the command reconstruction to obtain the restricted yaw moment constraint command;

[0071] The system reads the communication data dictionary of the yaw control channel and locates the physical start address of the 32-bit floating-point torque constraint field in the yaw control command data structure. It then reads the yaw wind-resistant integral absolute parameter generated in the previous stage. In the controller's floating-point comparator, it performs a numerical comparison command between the yaw wind-resistant integral absolute parameter and the limiting yaw moment boundary value. When the value of the integral absolute parameter is greater than or equal to the limiting yaw moment boundary value, a direct memory write operation is triggered, forcibly overwriting the value of the limiting yaw moment boundary value in single-precision floating-point format into the memory sector corresponding to the original integral parameter's input channel. When the value of the integral absolute parameter is less than the limiting yaw moment boundary value, the original integral absolute parameter is written to the channel. The system configures the data transmission bit width for direct memory access to word alignment mode and strictly verifies whether the overall byte length of the written control command data structure matches the preset 64-byte standard control frame length. Data bus alignment and command reconstruction are completed by padding with 0x00 null bytes at the end, resulting in a restricted yaw moment constraint command that is strictly constrained by dynamic boundaries and has a standard data flow specification.

[0072] Please see Figure 4 The specific steps of S3 are as follows:

[0073] S311: Extract the original sign bit identifier of the yaw angle error integral parameter, split the sign bit and the value bit and generate the sign bit identifier sequence, detect the structure of the restricted yaw moment constraint command data frame and locate the write-back overwrite storage address field, append the sign bit identifier sequence to the restricted yaw moment constraint command torque value field and perform sign composite operation to generate the sign composite yaw moment value.

[0074] The system addresses the address pointer in the dynamic memory area used to store the data split in the first stage, retrieving the previously separated and cached 8-bit unsigned integer sign identifier sequence. It iterates through each sign identifier in this sequence, loading it into register A. The protocol parser reads the data frame of the constrained yaw moment constraint instruction generated in the previous stage, scanning the frame header identifiers 0xAA and 0xBB, and sequentially offsetting to the data payload area containing bytes 12 to 15, precisely locating the write-back overwrite storage address field used to store the floating-point moment value. The arithmetic logic unit is activated, loading the unsigned moment value in this field into register B. Hardware-level branching logic is executed based on the sign identifier in register A: if the sign identifier is 1 (representing a negative error attribute in the original integral), a floating-point inversion instruction is called to convert the unsigned constrained moment in register B into a negative constrained yaw moment; if the sign identifier is 0, the original binary form of the value in register B is preserved. The numerical value after the symbolic composite operation is performed is latched into the result register to generate a symbolic composite yaw moment value that has both the original error correction direction and the current dynamic safety boundary.

[0075] S312: Based on the symbolic composite yaw moment value, read the restricted yaw moment constraint instruction storage address mapping table, perform a fixed-point write operation on the write-back overwrite storage address field and verify the consistency of the address segment check code, complete the data frame overwrite replacement and keep the field byte length unchanged, and generate a write-back yaw moment control frame.

[0076] A read addressing command is sent to the flash memory controller to load the underlying memory address mapping table of the constrained yaw moment constraint command. The key-value pairs in the mapping table are parsed to extract the absolute base address 0x0800C000 of the physical sector memory to be written to. The fixed-point write mask configuration of the advanced microcontroller bus architecture is configured, and write permissions to the memory protection unit are enabled. The symbolic composite yaw moment value is hard-written byte by byte through the data bus in a low-order-first-high-order memory end-order pattern to overwrite the memory address field. At the moment the write operation is completed, the hardware cyclic redundancy check coprocessor is activated. The entire 64-byte frame of the constrained yaw moment constraint command is captured, and a 16-bit cyclic redundancy check calculation is performed with the polynomial parameter set to 0x1021. After successful verification, a write protection latch command is sent to lock the length definition register of the current data block, maintaining the overall frame length at the absolute 64-byte specification, and generating a write-back yaw moment control frame that can be called by the lower-level mixer.

[0077] S313: Collect hovering lift values ​​of the UAV, detect the maximum thrust field of a single motor and perform a value substitution operation, write the hovering lift values ​​into the original maximum thrust field of a single motor and establish a mapping relationship between the thrust values ​​and the written-back yaw torque control frame, reconstruct the drive output based on the ratio of the thrust values ​​and the torque values, and generate the target wind-resistant thrust drive command.

[0078] Connect the data interfaces of the high-precision barometric altimeter and the Z-axis linear accelerometer on the UAV's central motherboard. Use a third-order Kalman filter to fuse the quadratic integral acceleration data and barometric altimeter data, and obtain the Z-axis vertical displacement deviation after removing the gravitational acceleration constant. Input the displacement deviation into the altitude maintenance proportional-integral-derivative controller, which outputs the overall hovering thrust that enables the UAV to overcome its own gravity and the downforce of the vertical wind field. Divide this overall hovering thrust by 4 (the total number of UAV motors) to obtain the single-axis hovering lift value under the current environment. Intercept and parse the reference thrust structure that the next-level attitude calculation matrix will call, forcibly erase the static single-motor maximum thrust field, and perform a fixed-point numerical substitution operation using the previously obtained hovering lift value. Construct a hybrid control mapping matrix algorithm model, using the hovering lift value as a reference column vector matrix, and convert the symbolic composite yaw torque value extracted from the write-back yaw torque control frame into a yaw torque offset row vector. Based on the multi-rotor control dynamics distribution equations, the reference lift is added to or subtracted from the corresponding torque distribution ratio (e.g., the thrust value of the right front rotor equals the hovering lift value plus the roll torque plus the pitch torque minus the write-back yaw torque), reconstructing the output terminal drive thrust values ​​of the four independent motors. The four independent thrust parameters are integrated with the frame header and frame tail encapsulation to generate the target wind-resistant thrust drive command that directly drives subsequent hardware actions.

[0079] Please see Figure 5 The specific steps of S4 are as follows:

[0080] S411: Collect the UAV's preset thrust duty cycle data table, extract the pulse width modulation duty cycle parameters that match the target's wind-resistant thrust drive command, perform a search operation based on the thrust value in the target's wind-resistant thrust drive command, obtain the corresponding pulse width modulation duty cycle value, and generate a pulse width modulation duty cycle parameter sequence.

[0081] The system reads the pre-stored UAV thrust duty cycle data table from the read-only partition of the flash memory. The quantization process for this data table is as follows: In a calm wind dynamometer environment, an incremental pulse width modulation (PWM) control level from 1000 microseconds to 2000 microseconds is input to the brushless motor of the rotor assembly, with a step value of 10 microseconds. A high-frequency tension sensor collects the corresponding thrust physical quantity output value. A mean-law filtering algorithm is applied to each sampling point to remove mechanical vibration interference extreme values. Finally, the filtered thrust value is used as the index key, and the corresponding microsecond pulse width duration is used as the target value, thus constructing a nonlinear lookup table containing 100 nodes. The specific thrust values ​​of each motor contained in the target wind-resistant thrust drive command are extracted. A binary search method is applied to perform an index search in the data table for this thrust value. The left and right boundary index parameters of the array are initialized, and the thrust value corresponding to the midpoint index is compared with the target thrust value, continuously narrowing the search interval until the two adjacent upper and lower thrust nodes are locked. After obtaining the upper and lower nodes, the thrust physical quantities and pulse width in microseconds corresponding to the two nodes are extracted. Using a first-order linear interpolation polynomial algorithm, the exact microsecond-level duty cycle corresponding to the target thrust value is calculated. The calculated exact microsecond values ​​of the four channels are packaged and integrated into a one-dimensional array to complete the generation of the pulse width modulation duty cycle parameter sequence. The data structure of the above processing is shown in Table 2.

[0082] Table 2 Mapping Table of Target Wind-Resistant Thrust and Pulse Width Modulation Duty Cycle

[0083]

[0084] Table 2 lists the process parameters for finding the exact duty cycle using a first-order linear interpolation algorithm in the embodiments. This table demonstrates the process by which the data table lookup operation perfectly quantizes the target thrust value into the underlying level execution pulse width, and this quantization result is directly input into the drive circuit.

[0085] S412: Based on the pulse width modulation duty cycle parameter sequence, read the communication interface of the rotor motor electronic speed controller, and send it to the rotor motor electronic speed controller one by one according to the matching duty cycle value to generate motor speed control command.

[0086] Based on the generated pulse width modulation duty cycle parameter sequence, the comparison output channel parameters of the advanced general-purpose timer are configured. The timer's base counting clock is set to 72 MHz, the prescaler value is set to 71, and the auto-reload register value is set to 1999, thus configuring a rigorous 500 Hz reference control cycle. The general-purpose input / output port communication interface connected to the four rotor motor electronic speed controllers is read and activated. The duty cycle microsecond values ​​matched by each channel in the sequence are sequentially loaded into the four independent capture and comparison registers of the timer. The direct memory access controller is activated, bypassing the central processing unit core, to periodically push the updated comparison values ​​to the timer peripherals. The hardware timer uses a dynamic matching trigger mechanism between the counter value and the comparison register value to automatically generate high-low level transition waveforms with extremely high timing accuracy on the communication interface pins, i.e., square wave electrical signals corresponding to the duty cycles. These are sent one by one to the corresponding rotor motor electronic speed controller signal input terminals through physical twisted-pair wires to generate the hardware-level motor speed control commands.

[0087] S413: Receives the status confirmation signal from the rotor motor electronic speed controller to confirm that the motor speed control command has been successfully transmitted and executed, and generates the terminal motor hardware drive command by combining the motor feedback data.

[0088] The microcontroller's universal asynchronous transceiver receiver channel, configured to operate at 115,200 bits per second, is activated. High-frequency digital signals are received from the rotor motor's electronic speed controller via a separate single-wire telemetry interface. An automaton receiver state model is established, including frame header identification, load byte acquisition, and redundancy check. A continuous 10-byte data stream is captured, and the flag signal indicating that the speed controller has received the pulse width command and confirmed its response is extracted. Status confirmation signals from the load, such as the speed controller's internal insulated gate bipolar transistor temperature, bus voltage, and commutation frequency, are extracted. An XOR operation and an 8-bit cyclic redundancy check are performed on the received full-frame data, and the checksum is compared with the checksum at the end of the frame. After confirming the checksum matches, it is determined that the motor speed control command has been successfully transmitted at the electrical level without attenuation or distortion. The real-time commutation frequency fed back by the speed controller is divided by the number of motor pole pairs to calculate the actual mechanical speed. This feedback data is merged with the previously issued duty cycle command data and written into the command execution confirmation buffer, generating a terminal motor hardware driver command with closed-loop handshake authentication information.

[0089] Please see Figure 6 The specific steps of S5 are as follows:

[0090] S511: Based on the terminal motor hardware drive command, monitor the response signal of the wind-resistant rotor motor at the end of the UAV arm, collect the rotation status data of the rotor mechanism of the rotor motor, trigger the rotation action of the motor rotor mechanism, start the UAV attitude balance adjustment module, and generate a preliminary attitude adjustment signal.

[0091] The terminal motor hardware driver command is sent to the underlying hardware abstraction layer. The multi-channel voltage comparator peripheral with built-in digital isolator is enabled. The electromagnetic response signal inside the wind-resistant rotor motor at the end of the UAV arm is monitored at high frequency through the back EMF zero-crossing detection circuit of the three-phase stator winding. The zero-crossing signal is input to the phase-locked loop digital processing unit to calculate the rotor's current electrical angle orientation. The collected electrical angle change rate is extracted and converted into the rotor motor rotor mechanism's rotational angular velocity data in the spatial coordinate system. When the angular velocity value exceeds the set dead-zone start threshold, the driver core logic determines that the rotor mechanism's rotation action has been formally triggered. A hardware interrupt signal is then sent to forcibly initiate the UAV attitude balance adjustment. The quaternion update frequency of the three-axis gyroscope and three-axis accelerometer is initialized to 1000 Hz. The complementary filtering algorithm operator in the attitude calculation program is invoked. Based on the current rotor mechanism's feedback speed, the reference attitude matrix offset is recalibrated. A difference element dataset containing roll angle difference, pitch angle difference, and yaw rate difference is calculated and combined to generate a preliminary attitude adjustment signal as the initial compensation reference.

[0092] S512: Based on the initial attitude adjustment signal, the attitude of the UAV is corrected in real time. Combined with the rotor motor speed and thrust value, the attitude change is balanced and adjusted, and dynamic trajectory correction commands are generated.

[0093] A cascaded proportional-integral-derivative (PID) correction network is established, comprising an inner-loop angular velocity controller and an outer-loop angle controller. The initial attitude adjustment signal is introduced into the input node of the outer-loop angle controller to calculate the deviation between the desired UAV attitude angle and the real-time sensor-analyzed attitude angle. The outer-loop controller multiplies the deviation by a preset proportional gain matrix and outputs the desired angular velocity. This desired angular velocity is fed into the inner-loop angular velocity controller, and residual calculation is performed with the real-time angular velocity monitored by the gyroscope. The rotor motor speed (e.g., 4200 revolutions per minute) and thrust value (e.g., 15.6 Newtons) determined in the previous step are extracted as adaptive parameter tuning variables. A gain scheduling table mechanism based on motor speed is established: if the current motor speed exceeds 80% of the rated speed, a gain attenuation function is activated, reducing the inner-loop differential time constant by an attenuation coefficient of 0.9 to smooth out high-frequency oscillations caused by blade aeroelastic deformation at high speeds; if the rate of change of the motor thrust value exceeds 15 Newtons per second, the weight of the integral feedforward term is increased through nonlinear adjustment to eliminate steady-state drag errors caused by large maneuvers. The output of all inner and outer loop control quantities after speed and thrust balance adjustment is integrated and encapsulated into a three-dimensional spatial trajectory coordinate system to generate dynamic trajectory correction commands for fine-tuning the flight path offset.

[0094] S513: Combines trajectory correction commands and initial attitude adjustment signals, records dynamic adjustment data and stable state parameters during the control process, outputs control logs, and generates a record of UAV anti-wind disturbance stability control.

[0095] Initialize the serial peripheral interface and mount a secure digital storage card with a 32-bit file system formatted by the file allocation table. Allocate a 4096-byte direct memory access write operation double buffer block in memory. Combine and package the axis adjustment values ​​contained in the dynamic trajectory correction instructions and the pre-correction attitude angle deviations involved in the initial attitude adjustment signals. Real-time acquisition of system clock counter values ​​captures dynamic adjustment data such as inner and outer loop proportional-integral-derivative state parameters, actual output duty cycle, and mechanical speed calculated from back electromotive force during the control process. Convert this batch of dynamic data, along with the aforementioned relevant reference state parameters (such as rated speed reference value and single-axis hovering lift), into comma-separated value structured string data. Trigger the physical write command of the storage card in append write mode, continuously arranging and outputting the formatted strings to a write-only log text file. After completing the control cycle recording for the specified duration, close the file read / write pointers to generate a UAV anti-wind disturbance stability control record with full-process timing traceability capabilities. The specific format of the above data is shown in Table 3.

[0096] Table 3 Dynamic Stability Control Operation Log Data Table

[0097]

[0098] Table 3 lists the key sampling data nodes in the wind disturbance resistance stability control record output process.

[0099] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for wind-disturbed stability control of unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: S1: Monitor the historical cumulative yaw angle error integral parameters of the UAV airborne flight control processor's anti-wind disturbance yaw closed loop, perform absolute value extraction on the yaw angle error integral parameters, and generate yaw wind-resistant integral absolute parameters. S2: Collect the current wind-resistant roll channel torque and wind-resistant pitch channel torque of the UAV, calculate the equivalent horizontal thrust consumption and the boundary value of the ultimate yaw moment, select the boundary value of the ultimate yaw moment to replace the yaw wind-resistant integral absolute parameter, and output the constrained yaw moment constraint command. S3: Extract the original sign of the yaw angle error integral parameter and append it to the restricted yaw moment constraint command write-back overwrite storage address, collect the UAV hovering lift and replace the original single motor maximum thrust, and establish the target wind-resistant thrust drive command. S4: Collect the UAV preset thrust duty cycle data table, perform a search action according to the target wind-resistant thrust drive command, extract the corresponding pulse width modulation duty cycle parameters and send them to the rotor motor electronic speed controller, and obtain the terminal motor hardware drive command. S5: Based on the terminal motor hardware drive command, combined with the response drive signal of the wind-resistant rotor motor at the end of the UAV arm and the electronic speed controller, the rotor mechanism of the wind-resistant rotor motor is triggered to rotate, and the attitude balance state of the UAV is adjusted and the trajectory is corrected. The record of the UAV wind-resistant stability control is output. The specific steps for obtaining the constrained yaw moment command are as follows: S211: Collect the wind-resistant roll channel torque and wind-resistant pitch channel torque of the UAV, perform absolute value calculation on the torque of the two channels respectively, perform addition operation on the two absolute values ​​to form the total horizontal resultant torque, extract the UAV arm length and perform division operation on the total horizontal resultant torque to generate the equivalent horizontal thrust consumption value; S212: Based on the equivalent horizontal thrust consumption value, obtain the maximum thrust of a single motor, the reverse torque conversion parameter, the motor efficiency coefficient, the current motor speed, the rated speed reference value, and the air density reference value, and calculate the boundary value of the limiting yaw moment. S213: Based on the boundary value of the limiting yaw moment, perform a numerical replacement operation on the absolute integral parameter of the yaw wind resistance, write the boundary value of the limiting yaw moment into the moment constraint field of the yaw control channel and cover the original integral parameter input channel, keep the field length of the control command data structure consistent and complete the command reconstruction to obtain the restricted yaw moment constraint command; The formula for calculating the boundary value of the limiting yaw moment is: ； in, This represents the boundary value of the limiting yaw moment. This indicates the maximum thrust of a single motor. This represents the equivalent horizontal thrust consumption value. Indicates the inverse torque conversion parameter. This represents the motor efficiency coefficient. Indicates the current motor speed. This indicates the reference value for the rated speed.

2. The method for wind-resistant stability control of unmanned aerial vehicles according to claim 1, characterized in that: The calculation of equivalent horizontal thrust consumption specifically refers to extracting the absolute values ​​of the anti-wind roll channel torque and the anti-wind pitch channel torque of the UAV, adding them together, and dividing by the length of the UAV arm.

3. The method for wind-resistant stability control of unmanned aerial vehicles according to claim 1, characterized in that: The yaw wind resistance integral absolute parameters include the heading drift magnitude, attitude offset magnitude, and cumulative error magnitude. The constrained yaw moment constraint commands include output truncation threshold, dynamic compensation limit, and safety correction quota. The target wind resistance thrust drive commands include the desired speed reference, power distribution expectation, and lift adjustment ratio. The terminal motor hardware drive commands include waveform trigger pulse, current control level, and frequency adjustment sequence. The UAV wind disturbance stability control record includes attitude balance state scalar, trajectory correction feedback quantity, and hovering stability variables.

4. The method for wind-resistant stability control of unmanned aerial vehicles according to claim 1, characterized in that, The specific steps for obtaining the integral absolute parameter of yaw wind resistance are as follows: S111: Monitor the operation data stream of the anti-wind disturbance yaw closed loop inside the UAV's airborne flight control processor, extract the historical cumulative yaw angle error integral parameter value sequence, sort and verify the integral values ​​of each sampling period according to the timestamp order, and generate the yaw error integral sequence. S112: Based on the yaw error integral sequence, perform a sign bit splitting operation on each integral value, convert the positive and negative signs into independent sign identifier fields and retain the corresponding numerical fields, establish a mapping between the sign identifier fields and the numerical fields, and generate an integral sign numerical mapping set. S113: Based on the numerical mapping set of integral symbols, perform absolute value operation on the numerical fields, convert the signed integral value into an unsigned numerical value and retain the symbol identification field in an independent storage area, keep the correspondence of each time node unchanged and complete the data structure reorganization to generate the yaw wind resistance integral absolute parameter.

5. The method for wind-resistant stability control of unmanned aerial vehicles according to claim 1, characterized in that, The specific steps for obtaining the target wind-resistant thrust drive command are as follows: S311: Extract the original sign bit identifier of the yaw angle error integral parameter, split the sign bit and the value bit and generate a sign bit identifier sequence, detect the structure of the restricted yaw moment constraint command data frame and locate the write-back overwrite storage address field, append the sign bit identifier sequence to the restricted yaw moment constraint command torque value field and perform a sign composite operation to generate a sign composite yaw moment value. S312: Based on the symbolic composite yaw moment value, read the restricted yaw moment constraint instruction storage address mapping table, perform a fixed-point write operation on the write-back overwrite storage address field and verify the consistency of the address segment check code, complete the data frame overwrite replacement and keep the field byte length unchanged, and generate a write-back yaw moment control frame. S313: Collect the hovering lift value of the UAV, detect the maximum thrust field of a single motor and perform a value substitution operation, write the hovering lift value into the original maximum thrust field of a single motor and establish an association mapping relationship between the thrust value and the written-back yaw torque control frame, reconstruct the drive output based on the ratio of the thrust value and the torque value, and generate the target wind-resistant thrust drive command.

6. The method for wind-resistant stability control of unmanned aerial vehicles according to claim 1, characterized in that, The specific steps for obtaining the terminal motor hardware driver instruction are as follows: S411: Collect the UAV preset thrust duty cycle data table, extract the pulse width modulation duty cycle parameters that match the target wind-resistant thrust drive command, perform a search operation based on the thrust value in the target wind-resistant thrust drive command, obtain the corresponding pulse width modulation duty cycle value, and generate a pulse width modulation duty cycle parameter sequence. S412: Based on the pulse width modulation duty cycle parameter sequence, read the communication interface of the rotor motor electronic speed controller, and send it to the rotor motor electronic speed controller one by one according to the matching duty cycle value to generate motor speed control command; S413: Receive the status confirmation signal from the rotor motor electronic speed controller to confirm that the motor speed control command has been successfully transmitted and executed, and generate the terminal motor hardware drive command based on the motor feedback data.

7. The method for wind-resistant stability control of unmanned aerial vehicles according to claim 1, characterized in that, The specific steps for obtaining the UAV's wind-disturbance-resistant stability control record are as follows: S511: Based on the terminal motor hardware drive command, monitor the response signal of the wind-resistant rotor motor at the end of the UAV arm, collect the rotation status data of the rotor mechanism of the rotor motor, trigger the rotation action of the motor rotor mechanism, start the UAV attitude balance adjustment module, and generate a preliminary attitude adjustment signal. S512: Based on the preliminary attitude adjustment signal, the attitude of the UAV is corrected in real time, and the attitude change is balanced and adjusted by combining the rotor motor speed and thrust value, and a dynamic trajectory correction command is generated. S513: Combining the trajectory correction command and the initial attitude adjustment signal, record the dynamic adjustment data and stable state parameters during the control process and output the control log to generate a UAV anti-wind disturbance stability control record.

Images