Vibration control method and device of unmanned aerial vehicle, storage medium and equipment

Abstract

The invention discloses a vibration control method and device of an unmanned aerial vehicle, a storage medium and equipment, and the method comprises the steps: firstly obtaining a vibration signal generated by a sensor preset on a target unmanned aerial vehicle through the sensor; analyzing and processing the vibration signal, extracting vibration characteristic data, inputting the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal, inputting the fuzzy control signal into a PID (Proportion Integration Differentiation) controller, processing the fuzzy control signal in three modes of proportion, integration and differentiation to generate a PID control signal, and according to the PID control signal, controlling the PID controller to generate a PID control signal; and adjusting the damping force of a magnetorheological damper preset on the target unmanned aerial vehicle to control and reduce the vibration of the target unmanned aerial vehicle. According to the invention, the fuzzy controller is adopted to process the complexity and uncertainty of the vibration signal of the target unmanned aerial vehicle, and the PID controller is combined to realize the accurate control of the damping force of the magnetorheological damper, so that the vibration of the target unmanned aerial vehicle is reduced, and the flight and stealth performance is improved.

Description

Technical Field

[0001] The present application relates to the technical field of unmanned aerial vehicles, and in particular to a vibration control method, device, storage medium and equipment for unmanned aerial vehicles. Background Art

[0002] A drone is an unmanned aircraft. With the rapid development of drone technology, it has been widely used in various fields such as agriculture, logistics and transportation, and environmental monitoring. However, due to the light weight of drones and the high aerodynamic force in the flight environment, drones are prone to vibration and noise, which affects the flight and stealth performance of drones. Therefore, it is of great significance to effectively control the vibration of drones to improve their flight and stealth performance.

[0003] At present, when using the traditional passive suspension system for drone vibration control, although it can provide a certain buffer when the drone encounters airflow changes or maneuvering flight, it is difficult to effectively suppress the vibration of the drone. When using the active suspension system for drone vibration control, although it can provide higher adaptability and control accuracy to reduce the vibration of the drone, and to a certain extent improve the stability and controllability of the drone flight, it still has disadvantages such as high power consumption and high cost. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to provide a vibration control method, device, storage medium and equipment for a drone, which uses a fuzzy controller to process the complexity and uncertainty of the drone vibration signal, and combines a PID controller to achieve precise control of the damping force of the magnetorheological damper, so as to reduce the vibration of the drone by adjusting its damping force, thereby improving the stability and safety of the drone flight.

[0005] The present application provides a vibration control method for a drone, including:

[0006] Acquire a vibration signal generated by the target UAV through a sensor pre-set on the target UAV;

[0007] Analyzing and processing the vibration signal to extract vibration characteristic data, and inputting the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal;

[0008] The fuzzy control signal is input into a proportional-integral-differential (PID) controller, and the fuzzy control signal is processed by three methods: proportional, integral and differential to generate a PID control signal;

[0009] According to the PID control signal, the damping force of the magnetorheological damper pre-set on the target UAV is adjusted to control and reduce the vibration of the target UAV.

[0010] In a possible implementation, the vibration signal is an analog signal; and the step of analyzing and processing the vibration signal to extract vibration characteristic data includes:

[0011] Performing analog-to-digital conversion on the vibration signal to obtain a vibration digital signal corresponding to the vibration signal;

[0012] Performing filtering processing on the vibration digital signal to obtain the filtered vibration signal;

[0013] Normalizing the filtered vibration signal to obtain the normalized vibration signal;

[0014] The normalized vibration signal is subjected to spectrum analysis processing to extract vibration characteristic data corresponding to the vibration signal.

[0015] In a possible implementation, filtering the vibration digital signal to obtain the filtered vibration signal includes:

[0016] The vibration digital signal is filtered using a low-pass filter and / or a band-pass filter to obtain the filtered vibration signal.

[0017] In a possible implementation, performing spectrum analysis on the normalized vibration signal to extract vibration feature data corresponding to the vibration signal includes:

[0018] The normalized vibration signal is subjected to Fourier transform and / or wavelet transform processing to extract vibration characteristic data corresponding to the vibration signal.

[0019] In a possible implementation, after performing spectrum analysis on the normalized vibration signal to extract vibration characteristic data corresponding to the vibration signal, the method further includes:

[0020] Performing signal decomposition on the vibration characteristic data to obtain frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data;

[0021] The step of inputting the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal comprises:

[0022] Inputting the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data into a fuzzy controller for fuzzification processing to obtain fuzzy linguistic variables;

[0023] According to the pre-constructed fuzzy rules, the fuzzy language variables are processed by fuzzy logic operations to generate output values ​​of fuzzy control;

[0024] The output value of the fuzzy control is defuzzified to obtain a fuzzy control signal.

[0025] In a possible implementation, the fuzzy control signal is input into a proportional-integral-differential (PID) controller, and the fuzzy control signal is processed by three methods, namely, proportional, integral and differential, to generate a PID control signal, including:

[0026] Normalizing and filtering the fuzzy control signal to obtain a processed fuzzy control signal;

[0027] The processed fuzzy control signal is input into the PID controller, and the linear part of the input signal is processed by three methods: proportional, integral and differential to generate a PID control signal.

[0028] In a possible implementation, adjusting the damping force of the magnetorheological damper pre-set on the target UAV according to the PID control signal includes:

[0029] Converting the PID control signal into a control signal required by a magnetorheological damper pre-set on the target UAV;

[0030] The current or hydraulic pressure input to the magnetorheological damper is adjusted according to the control signal to change the magnetorheological characteristics of the fluid in the magnetorheological damper so as to adjust the damping force generated by the magnetorheological damper.

[0031] In a possible implementation, converting the PID control signal into a control signal required by a magnetorheological damper pre-set on the target UAV includes:

[0032] Using a pre-built mathematical model between the damping force of the magnetorheological damper and the input current, the PID control signal is converted into a target input current input to the magnetorheological damper pre-set on the target UAV;

[0033] According to the target input current, the input current of the magnetorheological damper is adjusted by using a current control circuit as a control signal required by the magnetorheological damper.

[0034] In a possible implementation manner, the installation position of the sensor pre-set on the target UAV includes at least one of a rotor root position, a position near an engine, and a fuselage center position.

[0035] The embodiment of the present application also provides a vibration control device for an unmanned aerial vehicle, comprising:

[0036] An acquisition unit, configured to acquire a vibration signal generated by the target UAV through a sensor pre-set on the target UAV;

[0037] An extraction unit is used to analyze and process the vibration signal, extract vibration characteristic data, and input the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal;

[0038] A generating unit, used for inputting the fuzzy control signal into a proportional-integral-differential (PID) controller, processing the fuzzy control signal in three ways, namely, proportional, integral and differential, to generate a PID control signal;

[0039] The control unit is used to adjust the damping force of the magnetorheological damper pre-set on the target UAV according to the PID control signal to control and reduce the vibration of the target UAV.

[0040] In a possible implementation, the vibration signal is an analog signal; and the extraction unit includes:

[0041] A first conversion subunit is used to perform analog-to-digital conversion on the vibration signal to obtain a vibration digital signal corresponding to the vibration signal;

[0042] A filter subunit, used for filtering the vibration digital signal to obtain the filtered vibration signal;

[0043] A normalization subunit, used for performing normalization processing on the filtered vibration signal to obtain the normalized vibration signal;

[0044] The spectrum analysis subunit is used to perform spectrum analysis on the normalized vibration signal to extract vibration characteristic data corresponding to the vibration signal.

[0045] In a possible implementation, the filtering subunit is specifically used for:

[0046] The vibration digital signal is filtered using a low-pass filter and / or a band-pass filter to obtain the filtered vibration signal.

[0047] In a possible implementation, the spectrum analysis subunit is specifically used for:

[0048] The normalized vibration signal is subjected to Fourier transform and / or wavelet transform processing to extract vibration characteristic data corresponding to the vibration signal.

[0049] In a possible implementation, the extracting unit further includes:

[0050] A signal decomposition subunit, used to perform signal decomposition on the vibration characteristic data to obtain frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data;

[0051] The extraction unit also includes:

[0052] A fuzzification subunit, used for inputting the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data into a fuzzy controller for fuzzification processing to obtain fuzzy linguistic variables;

[0053] A first generating subunit is used to perform fuzzy logic operation processing on the fuzzy language variable according to a pre-constructed fuzzy rule to generate an output value of fuzzy control;

[0054] The defuzzification subunit is used to perform defuzzification processing on the output value of the fuzzy control to obtain a fuzzy control signal.

[0055] In a possible implementation, the generating unit includes:

[0056] A processing subunit, used for normalizing and filtering the fuzzy control signal to obtain a processed fuzzy control signal;

[0057] The second generating subunit is used to input the processed fuzzy control signal into the PID controller, process the linear part of the input signal by three methods of proportion, integration and differentiation, and generate a PID control signal.

[0058] In a possible implementation, the control unit includes:

[0059] A second conversion subunit, used to convert the PID control signal into a control signal required by the magnetorheological damper pre-set on the target UAV;

[0060] The adjustment subunit is used to adjust the current or hydraulic pressure input to the magnetorheological damper according to the control signal, change the magnetorheological characteristics of the fluid in the magnetorheological damper, and adjust the damping force generated by the magnetorheological damper.

[0061] In a possible implementation, the second conversion subunit is specifically used for:

[0062] Using a pre-built mathematical model between the damping force of the magnetorheological damper and the input current, the PID control signal is converted into a target input current input to the magnetorheological damper pre-set on the target UAV;

[0063] According to the target input current, the input current of the magnetorheological damper is adjusted by using a current control circuit as a control signal required by the magnetorheological damper.

[0064] In a possible implementation manner, the installation position of the sensor pre-set on the target UAV includes at least one of a rotor root position, a position near an engine, and a fuselage center position.

[0065] The embodiment of the present application also provides a vibration control device for a drone, including: a processor, a memory, and a system bus;

[0066] The processor and the memory are connected via the system bus;

[0067] The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes any one of the implementation methods of the vibration control method of the above-mentioned drone.

[0068] An embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes any one of the implementation methods of the above-mentioned vibration control method for a drone.

[0069] The embodiment of the present application further provides a computer program product. When the computer program product is executed on a terminal device, the terminal device executes any one of the implementations of the above-mentioned vibration control method for a drone.

[0070] The embodiments of the present application provide a vibration control method, device, storage medium and equipment for an unmanned aerial vehicle. First, a vibration signal generated by a target unmanned aerial vehicle is obtained through a sensor pre-installed on the target unmanned aerial vehicle; then the vibration signal is analyzed and processed to extract vibration characteristic data, and the vibration characteristic data is input into a fuzzy controller for processing to obtain a fuzzy control signal. Then, the fuzzy control signal is input into a PID controller, and the fuzzy control signal is processed in three ways: proportional, integral and differential to generate a PID control signal. Then, according to the PID control signal, the damping force of a magnetorheological damper pre-installed on the target unmanned aerial vehicle is adjusted to control and reduce the vibration of the target unmanned aerial vehicle.

[0071] It can be seen that since the present application uses a fuzzy controller to process the complexity and uncertainty of the vibration signal of the target UAV, and combines it with the PID controller to achieve precise control of the damping force of the magnetorheological damper, the vibration of the target UAV can be reduced by adjusting the damping force generated by the magnetorheological damper, thereby achieving more precise, smarter and more economical vibration control of the target UAV, and also improving the flight stability and safety of the target UAV. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0073] Figure 1 A schematic diagram of a flow chart of a vibration control method for a drone provided in an embodiment of the present application;

[0074] Figure 2 A schematic diagram of the composition of a vibration control device for an unmanned aerial vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0075] Due to the light weight of UAVs and the large aerodynamic environment in which they fly, the vibration level of UAVs is high, which affects the flight performance and stealth performance of UAVs. Therefore, it is very important to effectively control the vibration of UAVs to improve their flight performance and stealth performance.

[0076] At present, the following four methods can usually be used for vibration control of drones: The first is to use a passive suspension system to control the vibration of drones, specifically using traditional mechanical components such as springs and dampers to absorb vibration energy. The advantage of this control method is that the control system has a simple structure and low cost. The disadvantage is that it cannot adapt to changes in different vibration frequencies and amplitudes, and the control effect is limited. The second is to use an active suspension system to control the vibration of drones, specifically using sensors to monitor the vibration state in real time, and using actuators to actively apply reverse force to offset the vibration. The advantages of this control method are high control accuracy and strong adaptability. The disadvantages are that the control system is complex, expensive, and consumes a lot of energy. The third is to use intelligent control algorithms to control the vibration of drones, specifically combining proportional-integral-differential (PID) control, fuzzy control, neural network and other algorithms to optimize the performance of the vibration control system. The advantages of this control method are that it can improve control accuracy and strong adaptability. The disadvantage is that it requires a lot of data training and debugging, and the algorithm design is relatively complex. The fourth is to use magnetorheological technology to control the vibration of drones, specifically using the characteristics of magnetorheological materials to adjust the damping force by changing the magnetic field strength to achieve vibration control. The advantages of this control method are fast response speed, high control accuracy and low energy consumption. The disadvantages are high cost of magnetorheological materials and complex control system design.

[0077] It can be seen that the existing methods for realizing UAV vibration control all have their own obvious shortcomings, such as insufficient control accuracy, high cost, high energy consumption and other problems, making it difficult to achieve the ideal control effect.

[0078] To solve the above-mentioned defects, the present application provides a vibration control method for a UAV. First, a vibration signal generated by the target UAV is obtained through a sensor pre-installed on the target UAV; then the vibration signal is analyzed and processed to extract vibration characteristic data, and the vibration characteristic data is input into a fuzzy controller for processing to obtain a fuzzy control signal. Then, the fuzzy control signal is input into a PID controller, and the fuzzy control signal is processed in three ways: proportional, integral and differential to generate a PID control signal. Then, according to the PID control signal, the damping force of the magnetorheological damper pre-installed on the target UAV is adjusted to control and reduce the vibration of the target UAV.

[0079] It can be seen that since the present application uses a fuzzy controller to process the complexity and uncertainty of the vibration signal of the target UAV, and combines it with the PID controller to achieve precise control of the damping force of the magnetorheological damper, the vibration of the target UAV can be reduced by adjusting the damping force generated by the magnetorheological damper, thereby achieving more precise, smarter and more economical vibration control of the target UAV, and also improving the flight stability and safety of the target UAV.

[0080] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0081] First embodiment

[0082] See also Figure 1 , is a flow chart of a vibration control method for a drone provided in this embodiment, the method comprising the following steps:

[0083] S101: Obtain a vibration signal generated by the target UAV through a sensor pre-set on the target UAV.

[0084] In this embodiment, any drone data for which vibration control is performed using this embodiment is defined as a target drone. In addition, in order to improve the vibration control effect on the target drone, the present application pre-installs sensors at different positions on the target drone to obtain vibration signals generated by the target drone at corresponding positions, so as to perform the subsequent step S102.

[0085] It should be noted that, considering that the installation position of the sensor is crucial to the accurate acquisition of vibration signals, the location close to the main vibration source of the drone should be selected as much as possible, and the sensor should be avoided from being installed in areas that may be disturbed, such as places with large air turbulence. In addition, the sensor location can be selectively installed according to the specific drone mission and requirements, but in order to improve the comprehensiveness and accuracy of the vibration signals obtained, it is usually necessary to install sensors in at least two different locations to compare and comprehensively analyze the vibration conditions of the target drone.

[0086] Specifically, an optional implementation method is that the installation position of the pre-set sensor on the target UAV may include but is not limited to at least one of the rotor root position, the position near the engine and the center of the fuselage. Among them, the rotor root position refers to the connection between the rotor and the fuselage. The vibration signal generated here is strong and can accurately reflect the working state of the rotor. It can directly reflect the rotor vibration and help to perform more accurate vibration control on the target UAV. Installing a sensor near the engine can directly capture the vibration caused by the target UAV engine, which helps to monitor and control the vibration source related to the engine, so as to detect and deal with abnormal vibration caused by the engine as early as possible. The sensor installed at the center of the fuselage is used to capture the overall vibration of the target UAV and assist sensors at other positions in data correction and comprehensive analysis. The global vibration data obtained by the sensor installed at this position can help with comprehensive analysis and vibration optimization of the entire machine.

[0087] In addition, in addition to installing sensors at key locations such as the rotor root, near the engine, and in the center of the fuselage to ensure that the vibration signals generated by the tiny target drone can be captured and processed. Sensors can also be installed at other locations of the target drone. For example, the sensor can also be installed at the end of the wing of the target drone to detect the transmission of vibrations caused by the rotor to the wing. Alternatively, the sensor can also be installed at the tail position of the target drone to capture the overall balance and vibration state of the fuselage. Alternatively, the sensor can also be installed at the landing gear position of the target drone to sense impact vibrations during takeoff and landing. Alternatively, the sensor can also be installed near the onboard equipment of the target drone to monitor micro-vibrations caused by the operation of the equipment.

[0088] It should be noted that the present application does not limit the specific types and components of the sensors pre-installed on the target drone. For example, a high-precision acceleration sensor (i.e., a high-sensitivity sensor) can be pre-installed at the corresponding position of the target drone to ensure that the tiny vibrations generated at the corresponding position can be captured. For example, the pre-installed sensor can be a 3-axis accelerometer with high resolution (13 bits) that can measure accelerations within the range of ±16g, which can be used to capture tiny vibration signals, especially suitable for the rotor root and the center of the fuselage. Alternatively, it can also be an ultra-low power 3-axis accelerometer with high precision and low noise characteristics, suitable for long-term vibration monitoring, can be used near the engine, and adapt to complex vibration environments. Alternatively, it can also be a 3-axis accelerometer that supports high-resolution mode and can capture tiny vibrations, ranging from ±2g to ±16g, which is particularly suitable for fine capture of tiny changes and is suitable for multiple drone positions. When installing the sensor, these high-precision acceleration sensors should be preferentially installed at the three key positions of the rotor root, near the engine, and the center of the fuselage to ensure that tiny vibration signals can be captured and processed.

[0089] In addition, gyroscope sensors can be selected to work with accelerometers to capture angular velocity and attitude changes. Alternatively, environmental sensors, such as temperature and pressure sensors, help calibrate and compensate for environmental effects on vibration data, but these sensors are usually only used as auxiliary and cannot replace high-precision accelerometers. High-precision accelerometers should be installed first in the three key locations mentioned above, and other types of sensors can be used as auxiliary supplements according to needs.

[0090] In addition, it should be noted that the present application does not limit the data type of the vibration signal generated by the target UAV. For example, the vibration signal can be data output in the form of an analog signal. For example, the vibration signal generated by the target UAV can be a 20Hz sinusoidal waveform vibration signal with an amplitude of 1.5V collected by a sensor installed at the root of the rotor; it can also be a 50Hz composite vibration signal collected by a sensor installed near the engine, with a main frequency of 50Hz, superimposed with 0.2V high-frequency noise, and an overall signal amplitude of 1.8V; it can also be a 10Hz low-frequency vibration signal with an amplitude of 1.2V collected by a sensor installed at the center of the fuselage.

[0091] S102: Analyze and process the vibration signal, extract vibration characteristic data, and input the vibration characteristic data into the fuzzy controller for processing to obtain a fuzzy control signal.

[0092] In this embodiment, the sensor pre-set on the target drone is used through step S101. After the vibration signal generated by the target drone is obtained, in order to improve the vibration control effect on the target drone, the vibration signal can be further analyzed and processed using an appropriate algorithm, such as a filtering algorithm, which is used to remove the noise of the vibration signal and smooth the data. Then, by analyzing the spectral characteristics of the vibration signal data, the vibration characteristic data related to the vibration of the target drone can be more accurately identified. The vibration characteristic data is then input into the fuzzy controller to process the nonlinear and highly uncertain parts, and a fuzzy control signal is obtained to execute the subsequent step S103.

[0093] When analyzing and processing vibration signals, it is necessary to ensure that the selected algorithm can ensure that the extracted vibration feature data is accurate and reliable. This involves adjusting and optimizing the algorithm parameters, as well as monitoring and verifying the data quality. Only by ensuring the accuracy and reliability of the data can the vibration problem of the target drone be effectively discovered, and corresponding measures can be taken in a timely manner to solve it.

[0094] Specifically, an optional implementation method is that when the vibration signal is an analog signal; the implementation process of "analyzing and processing the vibration signal to extract vibration characteristic data" in this step S102 may include the following steps S1021-S1024:.

[0095] S1021: Perform analog-to-digital conversion on the vibration signal to obtain a vibration digital signal corresponding to the vibration signal.

[0096] In this implementation, when the vibration signal is data output by the sensor in the form of an analog signal, it is first necessary to use an analog-to-digital converter (ADC) to convert the vibration signal into a digital signal and ensure data accuracy in order to execute the subsequent step S1022.

[0097] For example: Assuming that the vibration signal is a 20Hz sinusoidal vibration signal collected by a sensor installed at the root of the rotor, with an amplitude of 1.5V, when performing analog-to-digital conversion, assuming that the analog-to-digital converter (ADC) used has a 10-bit resolution (1024 levels) and an operating voltage of 3.3V, the digital result obtained after the signal conversion is (1.5V / 3.3V)*1024≈466. That is, the vibration digital signal corresponding to the vibration signal is a 20Hz sine wave with an amplitude of about 466.

[0098] Alternatively, assuming that the vibration signal is a 50Hz composite vibration signal collected by a sensor installed near the engine, with a main frequency of 50Hz, superimposed with 0.2V high-frequency noise, and an overall signal amplitude of 1.8V, when performing analog-to-digital conversion, assuming that the analog-to-digital converter (ADC) used is still 10-bit resolution (1024 levels), and the operating voltage is still 3.3V, the digital result obtained after the signal conversion is (1.8V / 3.3V)*1024≈558. That is, the vibration digital signal corresponding to the vibration signal is a main vibration signal of 50Hz, superimposed with slight high-frequency noise, and an amplitude of about 558.

[0099] Alternatively, assuming that the vibration signal is a 10Hz low-frequency vibration signal collected by a sensor installed at the center of the fuselage, with an amplitude of 1.2V, when performing analog-to-digital conversion, assuming that the analog-to-digital converter (ADC) used is still 10-bit resolution (1024 levels), and the operating voltage is still 3.3V, the digital result obtained after the signal conversion is (1.2V / 3.3V)*1024≈372. That is, the vibration digital signal corresponding to the vibration signal is a 10Hz low-frequency vibration signal with an amplitude of about 372.

[0100] In this way, the analog signal is converted into a discrete digital signal to form a digital signal that can be processed in subsequent steps.

[0101] S1022: Filter the vibration digital signal to obtain a filtered vibration signal.

[0102] In this implementation, after the vibration digital signal corresponding to the vibration signal is obtained through step S1021, the existing or future filtering algorithm can be further used to filter the vibration digital signal, remove noise and smooth data, and obtain a filtered vibration signal to perform subsequent step S1023. For example, a low-pass filter can be used to filter the vibration digital signal to remove high-frequency noise and retain the effective vibration signal, and / or a band-pass filter can be used to filter the vibration digital signal (usually refers to a signal in a specific frequency range) to eliminate irrelevant frequency interference.

[0103] It should be noted that when there is a lot of high-frequency noise in the vibration signal, a low-pass filter can be used for filtering. It is particularly suitable for processing vibration signals from the rotor root or near the engine, which may be affected by high-frequency noise, such as interference from electronic equipment or wind. The purpose of using a low-pass filter for filtering is to retain the effective low-frequency vibration signals, which are usually related to the overall vibration state and structural response of the drone.

[0104] When you need to focus on vibrations in a specific frequency range, you can use a bandpass filter for filtering. For example, for the vibration signal collected by the sensor installed in the center of the fuselage, if you want to exclude low-frequency structural swings and high-frequency electronic noise and focus on a specific frequency band (such as the operating frequency band of the engine or rotor), you can use a bandpass filter. The purpose is to exclude interference from irrelevant frequencies and extract effective vibration signals within a specific frequency range, which may be directly related to the health status or operating efficiency of specific components.

[0105] Among them, this application does not limit the vibration signal within a specific frequency range, which is usually related to the operating frequency of various components of the drone.

[0106] For example: Taking the rotor operating frequency as an example, the frequency range is 50Hz-200Hz. The bandpass filter can retain the signals within this frequency range, which can specifically analyze the balance state and rotation efficiency of the rotor and exclude noise and interference from other frequencies. Taking the engine vibration as an example, the frequency range is 100Hz-500Hz. The bandpass filter can retain the vibration signal of the corresponding frequency to evaluate the operating condition of the engine and identify potential faults or signs of performance degradation. Taking the structural resonance frequency as an example, the frequency range is 5Hz-20Hz. The bandpass filter can extract the signal within this frequency band to monitor the health of the overall structure of the drone and identify dangerous factors that may cause resonance. It can be seen that the use of bandpass filters for filtering can effectively separate the vibration signals within these specific frequency ranges, making subsequent analysis more accurate. By analyzing the vibration signals within these frequency ranges, the operating status and component health of the drone can be accurately monitored so that timely adjustments can be made under abnormal circumstances.

[0107] In addition, the present application also adopts an adaptive filtering algorithm to dynamically adjust the filtering parameters according to the changes in the real-time vibration signal to improve the filtering effect. The specific adjustment process includes: 1. Initialize the filter parameters: set the initial parameters, for example, set the initial filter coefficient or bandwidth range. 2. Real-time signal acquisition: receive the vibration data stream from the sensor in real time. 3. Calculate the error: use the current filter parameters to predict the signal and compare it with the actual received signal. Determine the error between the predicted value and the actual value. 4. Update the filter parameters: adjust the filter parameters according to the calculated error. For example: Increase the filter bandwidth: If the high-frequency noise in the signal is strong, the filter bandwidth may need to be increased to capture more valid signals. If the signal has too many low-frequency components, the filter bandwidth may need to be reduced to remove unnecessary low-frequency noise. Alternatively, dynamically adjust the coefficient: use an algorithm such as the least mean square error (Least Mean Squares, LMS) to adjust the filter coefficient to reduce the error. 5. Apply the adjusted filter to filter the newly collected signal using the adjusted filter parameters. Continue to monitor the filtering effect to ensure that the error is minimized. 6. Loop adjustment: Continuously update and repeat the above process to dynamically adapt to signal changes and optimize the filtering effect.

[0108] For example: Assume that the cutoff frequency of the low-pass filter is set to 50Hz. When the high-frequency noise in the signal is detected to increase, the corresponding adjustment process is: if it is found that the signal after filtering still contains a lot of high-frequency noise, the error calculation is performed to adjust the cutoff frequency to 100Hz to improve the suppression of high-frequency noise. After the adjustment, the signal noise is reduced and the signal quality is improved. In this way, through this adaptive adjustment, the filter can better adapt to the dynamically changing signal environment and provide more accurate vibration signal processing results.

[0109] S1023: performing normalization processing on the filtered vibration signal to obtain a normalized vibration signal.

[0110] In this implementation, after obtaining the filtered vibration signal through step S1022, in order to further convert the vibration signal into a form that can be processed by the fuzzy controller, an existing or future normalization algorithm, such as minimum-maximum normalization or Z-score normalization method, can be used to normalize the filtered vibration signal, and the filtered vibration signal data is linearly converted to between 0 and 1 to eliminate the dimensional influence brought by different sensors and measurement units, so that the vibration feature data extracted in the subsequent steps is more accurate.

[0111] For example, if the filtered vibration signal range is [10,50], after the minimum-maximum normalization, 10 can be converted to 0, 50 can be converted to 1, and other values ​​can be converted proportionally. Alternatively, using Z-score normalization, if the original data has a mean of 30 and a standard deviation of 5, the value 30 can be converted to 0, and other values ​​can be adjusted proportionally.

[0112] S1024: Perform spectrum analysis on the normalized vibration signal to extract vibration feature data corresponding to the vibration signal.

[0113] In the present implementation, after the normalized vibration signal is obtained through step S1023, in order to subsequently convert the vibration signal into a form that can be processed by the fuzzy controller, the normalized vibration signal can be further subjected to spectral analysis processing using existing or future spectral analysis methods. For example, the normalized vibration signal can be subjected to Fourier transform and / or wavelet transform processing to extract vibration characteristic data corresponding to the vibration signal for executing subsequent steps.

[0114] It should be noted that Fourier transform refers to the analysis and processing of normalized vibration signals, with the purpose of converting them into frequency domain signals to show the amplitudes of different frequency components. For example, after Fourier transform, the normalized vibration signal in the range of [0,1] may show that the main frequency components are at 10Hz and 50Hz, indicating that the vibrations at these two frequencies are more significant. It is suitable for spectrum analysis of stable and periodic signals.

[0115] Wavelet transform is used for multi-resolution analysis, which can capture transient vibration characteristics more accurately and is suitable for non-stationary signals. For example, when analyzing the vibration signal of a target drone during takeoff, wavelet transform can be used to capture the instantaneous changes and non-stationary characteristics of the signal and reveal the frequency changes in different time periods. It is suitable for the analysis of signals with time-varying characteristics.

[0116] In addition to Fourier transform and / or wavelet transform, other spectrum analysis methods can also be used to perform spectrum analysis on the normalized vibration signal, such as short-time Fourier transform (STFT), Hilbert-Huang transform (HHT), adaptive filter, etc. Among them, short-time Fourier transform (STFT) is suitable for analyzing the frequency characteristics of the signal in a local time period. Hilbert-Huang transform (HHT) is used for the analysis of nonlinear and non-stationary signals, and is suitable for the processing of complex vibration signals. Adaptive filters are used for spectrum analysis that dynamically adjusts the signal, and are suitable for signals with large changes.

[0117] Furthermore, after extracting the vibration feature data corresponding to the vibration signal, a recognition model can be constructed by machine learning algorithms (such as support vector machines, neural networks) to identify abnormal vibration patterns. Specifically, the existing feature data is first trained using a machine learning algorithm (such as a support vector machine or a neural network). Among them, the way to use the support vector machine for training is to train the feature data with the label (normal / abnormal) to establish a classification model. The way to use the neural network for training is to build and train a network model (such as a feedforward neural network or a convolutional neural network) to learn the complex relationship between features and labels. Then use unseen data to test the trained recognition model to evaluate its accuracy and robustness. In this way, after converting the vibration feature data corresponding to the extracted vibration signal into a data input format that can be recognized by the trained model, it can be input into the trained recognition model to predict whether the target drone has an abnormal vibration pattern. During the recognition process, the threshold for abnormal judgment can be set according to the model output. When the model detects that the abnormal pattern exceeds the set threshold, it automatically triggers an alarm and notifies the operator or system to take measures, or automatically adjusts system parameters or executes protective measures.

[0118] For example: the vibration feature data corresponding to the extracted vibration signal may include the main frequency components and energy of the vibration. When the recognition model is generated by training with a support vector machine, the training data set is used to distinguish between normal and abnormal vibrations, and the classification is performed by constructing support vectors. For example, the support vector machine model can be trained with normal vibration and fault vibration data. In this way, after the vibration feature data corresponding to the extracted vibration signal is converted into a data input format that can be recognized by the trained support vector machine model, it can be transmitted to the support vector machine model in real time for analysis. The model outputs the abnormal probability, and it is judged as abnormal if it exceeds the set threshold. For example, when the vibration amplitude exceeds the set threshold, the system automatically issues an alarm to notify the operator. In this way, the drone can intelligently monitor the vibration state, automatically identify abnormalities and issue warnings, thereby improving safety and stability. When the recognition model is generated by training with a neural network, a convolutional neural network (CNN) can be used to process the vibration signal feature map, identify and classify abnormal patterns, and the subsequent recognition process will not be repeated.

[0119] It can be seen that this pattern recognition method can automatically detect and classify vibration patterns of target drones, providing early warning of potential problems. In addition, through intelligent anomaly detection, it can also timely warn and handle vibration anomalies, thereby reducing the risk of failure. That is, once an abnormal pattern is detected, the system will immediately issue an alarm to notify the operator or trigger automatic response measures. The system automatically adjusts the control parameters based on the warning information and performs real-time vibration regulation to prevent the abnormal situation from deteriorating. Therefore, through timely detection and processing, the probability of equipment failure can be reduced and the safety and stability of the drone can be ensured.

[0120] In addition, in order to improve the accuracy of vibration detection, after installing sensors at the rotor root, near the engine and the center of the fuselage of the target UAV for multi-point synchronous measurement, the accuracy of vibration detection can also be improved by data fusion. Specifically, after extracting the corresponding vibration feature data (frequency, amplitude and other features) from the real signal obtained by the sensor at each position. Further, weights can be assigned according to the reliability and importance of different sensor data to achieve weighted fusion of data. For example, the vibration signal generated by the rotor root may be more important in identifying rotor-related vibrations. Among them, the fusion algorithm used is not limited, and data fusion algorithms such as Kalman filtering and weighted averaging can be used to integrate feature data from different sources into a comprehensive vibration assessment result. The fused data is then used for further analysis, such as identifying abnormal patterns. It is used to provide more accurate vibration pattern recognition and fault diagnosis information.

[0121] For example: The vibration signal generated at the root of the rotor may show high-frequency vibration, which indicates that the rotation of the rotor is unbalanced. The vibration signal generated near the engine may show medium and low-frequency vibration, indicating that the engine is unstable. The vibration signal generated at the center of the fuselage combines the effects of the above two and provides an overall vibration trend. In this way, after extracting the vibration feature data (such as spectral features) corresponding to the vibration signal from each position. The weighted average method can be used to combine the features from the rotor root and near the engine with the vibration feature data (such as spectral features) corresponding to the vibration signal extracted from each position such as the center of the fuselage to obtain a comprehensive vibration feature. In this way, the fused data obtained can more accurately reflect the overall vibration situation and improve the accuracy of fault diagnosis. In this way, the impact caused by single sensor error or noise can be reduced, and more reliable vibration feature data can be obtained.

[0122] On this basis, an optional implementation method is that the final vibration feature data can also be input into the fuzzy controller for processing to adjust the control amount of the target drone according to the characteristics of the vibration signal, thereby effectively reducing the impact of vibration. Among them, the fuzzy controller uses fuzzy logic and fuzzy rules to process the input vibration feature data and output the corresponding fuzzy control signal generated. In this process, it is necessary to consider how to design appropriate fuzzy rules and how to optimize the performance of the controller. The vibration signal needs to be preprocessed first so that it can be converted into a form that can be processed by the fuzzy controller. This is mentioned in the above steps S1021-S1024. Analysis and processing steps such as normalization and feature extraction of the vibration signal are mentioned to ensure that the fuzzy controller can accurately capture the important information of the vibration signal. Fuzzy rules are usually generated based on expert experience or data analysis design and are used to convert input signals into corresponding control outputs. These rules can be implemented through fuzzy reasoning, in which fuzzy logic operations are used to process the relationship between fuzzy rules to determine the final control amount. When designing fuzzy rules, it is necessary to consider the impact of different vibration modes on the control amount and how to adjust the control output according to the intensity and frequency of the vibration signal. After the fuzzy controller is designed, simulation and debugging are required to verify its performance and stability. This may involve testing and analyzing the control effects under different vibration scenarios, as well as adjusting and optimizing the fuzzy rules and parameters to improve the performance and adaptability of the controller. The optimized fuzzy controller is applied to the actual UAV system and monitored and adjusted in real time. Through continuous feedback and optimization, it can be ensured that the fuzzy controller can effectively reduce the vibration impact of the UAV and improve its flight stability and safety.

[0123] Specifically, an optional implementation method is that after obtaining the vibration characteristic data, first, the vibration characteristic data can be decomposed to obtain the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data. The specific decomposition process is: Assume that in the spectrum analysis, the main frequency components of the frequency domain signal obtained are 50Hz and 150Hz. Then, when decomposing the signal, these frequency components can be extracted separately to further decompose the different vibration modes of the signal. For example, the 50Hz vibration is decomposed into basic sine waves, and the 150Hz vibration is decomposed into another group of sine waves. These decompositions help to identify vibration characteristics and patterns at different frequencies. It can be understood that since spectrum analysis can provide the frequency components and energy distribution of the signal. On this basis, through signal decomposition, these components can be further analyzed and concretized into recognizable patterns, which helps to understand and process the vibration characteristics more deeply.

[0124] Then, the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data can be input into the fuzzy controller for fuzzification processing to obtain fuzzy linguistic variables.

[0125] The specific content and format of the fuzzy linguistic variables are not limited. For example, the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data can be converted into low, medium, and high fuzzy linguistic variables.

[0126] Specifically, due to the main frequency components and vibration characteristics of the extracted vibration signal, it can provide a deep understanding of the signal. These characteristics help to understand the vibration pattern and provide basic data for the definition of fuzzy variables. For example, by extracting the main vibration frequencies, it can be determined which frequency components have a significant impact on the stability of the target drone. The signal decomposition process decomposes the complex vibration signal into simpler patterns, which helps to refine and analyze specific vibration characteristics. Signal decomposition can reveal multiple vibration modes or components in the signal, which is very important for accurately defining fuzzy variables because it enables the fuzzy controller to process different vibration modes more accurately.

[0127] In this way, by converting the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data into fuzzy language variables (such as low, medium, and high), complex numerical data can be converted into a form that is easy to process and interpret, so that the fuzzy controller can manage and control vibration more effectively. Among them, feature extraction and signal decomposition play a role in the steps before the definition of fuzzy variables. They provide a deep understanding and concretization of the signal, so that the fuzzy variable definition can more accurately reflect the vibration situation.

[0128] Then, according to the pre-constructed fuzzy rules, fuzzy logic operations can be performed on the fuzzy language variables to generate the output value of the fuzzy control.

[0129] It should be noted that different fuzzy sets can also be divided according to the characteristics of the vibration signal (such as defining the range and boundaries of the fuzzy set). In this way, the role of the obtained fuzzy set is: for the fuzzy controller, it can be used to control and adjust the vibration of the drone. The fuzzy set provides a linguistic description of the vibration state (such as low, medium, and high), enabling the controller to apply appropriate control strategies according to different vibration levels. At the same time, it can help identify abnormal vibration patterns and take appropriate measures to deal with them. In addition, control decisions can be made according to the definition of fuzzy sets during real-time control and adjustment, thereby optimizing the stability and performance of the drone.

[0130] Among them, fuzzy rules are usually formulated based on expert experience and data analysis, such as "if the vibration signal is high and the frequency is low, the control output is small". Various fuzzy rules can be used to build a fuzzy rule base, and through simulation and experiments, the rule base can be continuously adjusted and optimized to ensure the accuracy and effectiveness of the rules. The role of building a fuzzy rule base is to provide clear control rules for real-time adjustment of the vibration control strategy of the drone. And convert the vibration signal and frequency characteristics into control output to optimize the stability and flight performance of the target drone. And through fuzzy logic to handle uncertainty and incomplete information, improve the robustness of the system.

[0131] For the optimization of the fuzzy rule base, the specific implementation process can be as follows: the first step is to formulate preliminary rules, that is, to formulate preliminary fuzzy rules based on expert experience and historical data. For example, "if the vibration signal is high and the frequency is low, the control output is small", etc. The second step is to conduct simulation tests, that is, to test the preliminary rules in a simulated environment and evaluate their effect on vibration control. Record the test results and system response. The third step is to conduct data analysis, that is, to analyze the simulation and actual test data to identify the deficiencies and improvements in the rules. The fourth step is to adjust the rules, that is, to adjust the conditions and output values ​​of the fuzzy rules according to the analysis results. For example, refine the threshold in the rule or adjust the intensity of the control output. The fifth step is to perform iterative optimization, that is, to repeatedly simulate and experiment, and gradually optimize the rule base. After each adjustment, the performance of the rule base needs to be verified to ensure that it can effectively control vibration under different circumstances. The sixth step is to conduct actual application verification, that is, to verify the optimized rule base in the actual drone system, and further adjust it to cope with changes and challenges in the real environment. In the future, the fuzzy rule base can also be continuously adjusted and optimized through real-time vibration data feedback to adapt it to different vibration modes. This will improve the adaptability and robustness of the control system and enable it to cope with a variety of complex vibration situations.

[0132] On this basis, based on the set of defined fuzzy language variables and fuzzy rules contained in the fuzzy rule base. The process of processing and reasoning the input signal (i.e., fuzzy language variable). That is, the input signal (i.e., fuzzy language variable) is combined with the rules in the rule base to calculate and generate the output value of the corresponding fuzzy control. Among them, the role of the fuzzy rule base is to provide the basis of the fuzzy logic system, and the control strategy and response under different input conditions are defined by the rule base. And the practical application of the rule base is realized through fuzzy reasoning, and the input signal is mapped to the control output through fuzzy logic operation, so as to realize the real-time adjustment and optimization of the system. Among them, the application does not limit the reasoning method adopted by fuzzy reasoning, and can be selected according to the actual situation and experience value, such as the more commonly used reasoning methods such as Mamdani reasoning and Sugeno reasoning.

[0133] Finally, the output value of the fuzzy control may be defuzzified to obtain a fuzzy control signal for executing the subsequent step S103.

[0134] Among them, this application does not limit the fuzzy processing method used. For example, the maximum membership method, weighted average method, etc. can be used to convert the output value of the fuzzy control into an accurate control quantity as a fuzzy control signal, which is passed to the PID controller through the subsequent step S103.

[0135] In addition, various existing or future real-time enhancement methods can be used to optimize the calculation speed of the fuzzy control algorithm in real time to ensure that the fuzzy controller can respond to vibration changes in real time, including but not limited to: Optimizing the algorithm, that is, improving the calculation efficiency of the fuzzy control algorithm and reducing the processing time. For example, simplifying the rule base, using efficient data structures, and reducing computational complexity. Hardware acceleration, that is, using high-performance processors or dedicated hardware (such as FPGA, DSP, etc.) to accelerate algorithm operations and increase processing speed. Parallel computing, that is, assigning computing tasks to multiple processing units (such as multi-core CPUs or GPUs) to achieve parallel processing and shorten response time. Real-time scheduling, that is, ensuring that the control system prioritizes real-time tasks to avoid interference from other tasks. These measures can ensure that the fuzzy controller can quickly process vibration data and adjust the control output in a timely manner to improve the stability and responsiveness of the drone.

[0136] S103: Input the fuzzy control signal into the PID controller, process the fuzzy control signal through three methods of proportion, integration and differentiation, and generate a PID control signal.

[0137] In this embodiment, after the fuzzy controller is used to process the complexity and uncertainty of the vibration signal of the target UAV in step S102, the PID controller can be further used to process the linear part of the vibration signal of the target UAV, and the hybrid control strategy formed by the combination of the two can improve the overall control performance and response speed of the target UAV. Specifically, after the fuzzy control signal is obtained in step S102, the fuzzy control signal can be further input into the PID controller, and the fuzzy control signal can be processed in three ways: proportional, integral and differential to generate a PID control signal for executing the subsequent step S104.

[0138] Among them, the PID controller adjusts the output control quantity by comprehensively utilizing the information of the current state, historical state and expected state of the system to achieve stable control of the system. Before inputting the control quantity into the PID controller, it is necessary to ensure that the output of the fuzzy controller can effectively reflect the actual situation of the system and has optimized the control effect as much as possible. The fuzzy control signal output by the fuzzy controller is used as the input of the PID controller. In this way, the PID controller can adjust the control quantity according to the fuzzy control signal output by the fuzzy controller to further optimize the control effect of the target UAV. In this process, the PID controller processes the input signal (i.e., the fuzzy control signal) through three parts: proportional, integral and differential, and generates the corresponding control output (PID control signal). Among them, the proportional part is used to adjust the size of the control quantity according to the current error size; the integral part is used to eliminate the steady-state error of the entire control system; the differential part is used to predict future development trends to further adjust the size of the control quantity. The parameters of the PID controller need to be adjusted to adapt it to the characteristics of the fuzzy controller output. The parameters of the PID controller include proportional coefficient, integral time and differential time. The selection of these parameters is crucial to the control effect. By debugging and optimizing the parameters of the PID controller, it can be better adapted to the changes in the output of the fuzzy controller, thereby further improving the control performance of the entire system. After adjusting the parameters, simulation and experimental verification are required to evaluate the performance and stability of the PID controller. This may involve testing and analyzing the control effects under different control scenarios, as well as further adjusting and optimizing the parameters of the PID controller. Through continuous feedback and optimization, it can be ensured that the PID controller can effectively optimize the overall control effect and improve the flight stability and stealth performance of the target UAV.

[0139] Specifically, an optional implementation method is that after obtaining the fuzzy control signal, the fuzzy control signal can first be normalized and filtered, such as by a signal processing module or system, to obtain a processed fuzzy control signal, ensuring that the processed fuzzy control signal is smooth and noise-free.

[0140] For example: Assume that the data range of the input fuzzy control signal is -200 to 500. After normalization, the signal can be linearly converted to the range of 0 to 1. After filtering, the processed fuzzy control signal can be 0.143, 0.251, 0.372, 0.485, etc. Among them, the filtering process can use a low-pass filter to remove high-frequency noise and retain low-frequency signals. Assume that the filter is set to a cut-off frequency of 0.3Hz. After filtering, the signal can be smoothed to 0.15, 0.25, 0.35, 0.47, etc.

[0141] It can be seen that normalization and filtering can ensure that the input signal is within a uniform range and remove noise, making the signal smoother and more processable. Then, the processed fuzzy control signal can be input into the PID controller, and the linear part of the input signal is processed by three methods: proportional, integral and differential to generate a PID control signal.

[0142] It should be noted that the functions of the three parts of the PID controller in this step are as follows: The proportional (P) is used to adjust the control amount according to the current error size and quickly respond to the system error. For example, suppose the target altitude of the target drone is 1000 meters, the current altitude is 980 meters, and the error is 20 meters. If the proportional gain K p = 2, the control output is 20 × 2 = 40. The proportional controller will quickly increase the thrust to try to adjust the target drone's altitude to the target value. Integral (I) is used to accumulate historical errors, eliminate the system's steady-state error, and improve control accuracy. For example, if the drone hovers around 980 meters for a long time, even if the proportional controller has reduced the error, there may still be a small steady-state error. Integral gain K i is 0.1, the accumulated error is 20 meters, and the integral value is 20×0.1=2. Then the integral controller will adjust the control amount to eliminate these long-term errors and make the drone reach 1000 meters more accurately. The differential (D) is used to predict the error change trend, adjust the control amount in advance, and reduce overshoot and oscillation. For example, suppose the altitude of the drone is increasing rapidly and the error change rate is 10 meters / second. If the differential gain K d =1, the differential control amount is 10×1=10. The differential controller will reduce the thrust in advance to prevent the UAV from exceeding the target value and causing oscillation. In this way, the PID controller can effectively respond, adjust and optimize the control system of the target UAV through the combination of these three parts (P, I, D), and improve the control accuracy and stability.

[0143] It should also be noted that the present application does not limit the method for parameter optimization of the PID controller, and can be selected according to actual conditions and empirical values, such as the Ziegler-Nichols method can be used to determine the initial parameters, and the genetic algorithm can be used for parameter optimization, to search for the global optimal solution and improve the performance of the controller. According to the real-time feedback signal, the PID parameters are adaptively adjusted to improve the overall system adaptability. Among them, the implementation steps of the Ziegler-Nichols method are: first, the integral (I) and differential (D) gains are set to zero, and then the proportional (P) gain is gradually increased until the system begins to oscillate stably. Record the proportional gain and oscillation period at this time. Then use the Ziegler-Nichols formula to calculate the initial value of the PID parameters based on these data. The implementation steps of the genetic algorithm are: first define the fitness function, such as the overshoot of the system, the response time, etc. Then use the genetic algorithm to search for the optimal combination of PID parameters, and gradually approach the optimal solution through operations such as crossover and mutation. For example, using genetic algorithms to iterate the proportional, integral, and differential parameters multiple times, after optimization, it was found that the proportional gain was 1.2, the integral gain was 0.4, and the differential gain was 0.1, which can significantly reduce the overshoot and oscillation of the system and improve stability. The implementation steps of adaptive adjustment are: real-time monitoring of system performance and dynamic adjustment of PID parameters according to real-time errors. The algorithm adjusts the proportional, integral, and differential gains based on feedback to adapt to different operating conditions. For example, assuming that a large error occurs in the system, the integral gain may be increased to reduce the steady-state error. If the system responds excessively, the proportional gain is reduced or the differential gain is increased to reduce the overshoot. In this way, by monitoring the system status in real time and using the adaptive algorithm to dynamically adjust the PID parameters, it can be ensured that the controller can achieve optimal performance under different working conditions. This improves the adaptability and robustness of the control system, reduces the debugging time, and improves the stability of the target UAV in various complex environments.

[0144] In addition, considering the various performance indicators of the control system (such as steady-state error, response time, overshoot), multi-objective optimization can also be performed to determine the optimal PID parameter combination. This ensures that the control system can achieve better performance in multiple performance indicators and improve the control quality of the overall system. The specific implementation process of multi-objective optimization is as follows: First, define the optimization target: steady-state error, that is, the error of the control system when it reaches a steady state. Response time, that is, the time required for the system to reach a stable state from the input signal change. Overshoot, that is, the degree to which the system response exceeds the final steady-state value. Then, select the optimization method: objective function construction, such as defining steady-state error, response time and overshoot as objective functions. The objective function may be the weighted sum of these indicators. Optimization algorithm selection: Commonly used optimization algorithms include genetic algorithms, particle swarm optimization, simulated annealing, etc. Next, execute the optimization algorithm: generate a set of random combinations of initial PID parameters. Then use the objective function to evaluate the performance of each set of parameters. Adjust the PID parameters according to the rules of the optimization algorithm, and update the parameter combination to find the optimal solution. Repeat the above optimization process until the stop criterion is met (such as the convergence of the objective function value or the predetermined number of iterations). Then, we adjust the parameters and use the optimization results to apply the optimal PID parameters to the control system. We verify the effectiveness of the optimization results in the actual system and make further adjustments as needed.

[0145] Example: Objective function construction: Assume that the weights of steady-state error, response time and overshoot are set to 0.4, 0.3 and 0.3 respectively. The objective function can be expressed as: Objective function = 0.4 × steady-state error + 0.3 × response time + 0.3 × overshoot; Objective function = 0.4 × steady-state error + 0.3 × response time + 0.3 × overshoot; Optimization algorithm selection: Use genetic algorithm to initialize a set of random PID parameters (K p ,K i ,K d ), evaluate its performance in the objective function. Execute the optimization algorithm: generate 100 random combinations of PID parameters, calculate the value of each parameter in the objective function, generate new PID parameter combinations through the selection, crossover and mutation operations of the genetic algorithm, and after multiple iterations, find the PID parameter combination that minimizes the objective function value. Then use the optimization results to replace the optimal PID parameters (such as K p =1.5,K i ,=0.5,K d =0.1) is applied to the actual control system. In addition, these parameters are tested in the actual system to check whether they meet the design requirements and fine-tune according to the test results. Through this multi-objective optimization process, a balance can be achieved between multiple performance indicators to ensure that the PID controller achieves the best performance in various indicators.

[0146] In addition, the system output signal can be monitored in real time and fed back to the PID controller. The error calculation can be performed based on the feedback signal, that is, the error between the actual output and the expected output is calculated as the basis for the next step of control.

[0147] In addition, by optimizing the calculation speed and real-time performance of the PID control algorithm, the controller can be ensured to respond quickly to high-frequency vibration signals, which can significantly improve the real-time response capability of the control system and reduce the impact of vibration on the flight stability of the target UAV. This can be achieved through the following steps:

[0148] The first step is to simplify the algorithm:

[0149] Reduce the amount of calculation: Simplify the calculation process of the PID algorithm, such as using more efficient mathematical operations and reducing unnecessary calculation steps.

[0150] Optimize data structures: Use efficient data structures and algorithms to reduce memory access time and computational complexity.

[0151] The second step is efficient programming:

[0152] Optimize code: Use compiler optimization options and write efficient code to reduce redundancy and inefficient operations.

[0153] Leverage hardware acceleration: If possible, use hardware acceleration features (such as DSP or FPGA) to increase the speed of calculations.

[0154] The third step is the real-time operating system (RTOS):

[0155] Use RTOS: Run the PID control algorithm in a real-time operating system to ensure real-time scheduling and fast response.

[0156] Priority setting: Assign high priority tasks to the PID control algorithm to ensure that it has priority in obtaining computing resources in the system.

[0157] The fourth step is parallel processing:

[0158] Multithreading / multi-core processing: Utilize multithreading or multi-core processors to parallelize the calculation of PID control algorithms and improve processing speed.

[0159] Task decomposition: Decompose the PID control task into multiple subtasks and process different computing tasks in parallel.

[0160] The fifth step is sampling and updating frequency:

[0161] Improve sampling frequency: Increase the sampling frequency to capture vibration signals more frequently and improve the system's responsiveness to high-frequency signals.

[0162] Adjust Update Interval: Optimize the controller's update interval to ensure it is short enough to respond quickly to signal changes.

[0163] The sixth step is to optimize the calculation accuracy:

[0164] Use appropriate data types: Choose an appropriate data type (such as floating-point or fixed-point) to balance computational accuracy and speed.

[0165] Error Correction: Taking numerical errors into account in calculations and making necessary corrections to improve the accuracy of the algorithm.

[0166] For example:

[0167] The first step is to simplify the algorithm: use a simplified PID formula, such as avoiding complex mathematical functions or approximate calculations to reduce computational complexity.

[0168] The second step is efficient programming: write C or C++ code and use compiler optimization options (such as -O2 or -O3) to improve code execution efficiency.

[0169] The third step is real-time operating system: configure the high priority of the PID control task in the RTOS to ensure that it is given priority in system scheduling.

[0170] The fourth step is parallel processing: In a multi-core processor system, the PID control algorithm is split into multiple threads to handle different computing tasks respectively.

[0171] Step 5 Sampling and update frequency: Increase the sampling frequency of the vibration signal to several thousand times per second to capture high-frequency vibration signals.

[0172] The sixth step is to optimize the calculation accuracy: use fixed-point numbers for PID calculation to increase the operation speed while maintaining sufficient accuracy. Through these measures, the calculation speed and real-time performance of the PID control algorithm can be significantly improved, ensuring that the control system can quickly respond to high-frequency vibration signals, thereby improving the overall performance of the system.

[0173] S104: According to the PID control signal, the damping force of the magnetorheological damper pre-set on the target UAV is adjusted to control and reduce the vibration of the target UAV.

[0174] In this embodiment, after the PID control signal is generated by step S103, the control signal can be further output to a preset actuator to control the damping force generated by the magnetorheological damper. And by effectively adjusting the damping force, the vibration of the target drone is reduced and the stability and safety of its flight are improved. Among them, before outputting the control signal, the PID control signal output by the PID controller needs to be converted into a damping force suitable for the magnetorheological damper. To this end, first of all, it is necessary to design a suitable control interface to convert the output of the PID controller into the control signal required by the magnetorheological damper. This may involve circuit design and signal conversion technology to ensure that the control signal can be effectively transmitted to the magnetorheological damper and the damping force can be accurately adjusted. The converted control signal is then output to the magnetorheological damper to adjust the damping force it generates.

[0175] The present application does not limit the specific composition of the preset actuator, which refers to the device or component used to actually implement the control signal. Specifically in the application of magnetorheological damper, the actuator may include but is not limited to the following:

[0176] Electric actuator: An electric device that controls the magnetorheological damper and adjusts the damping force of the damper according to the PID control signal.

[0177] Servo motor: The servo motor that drives the magnetorheological damper adjusts the damping force by precisely controlling its rotation.

[0178] Pneumatic or hydraulic cylinders: In some systems, pneumatic or hydraulic cylinders can be used to adjust the damping force and make precise adjustments based on the control signal.

[0179] These actuators convert control signals into actual physical operations, adjusting the damping characteristics of the magnetorheological dampers, thereby effectively controlling the vibration of the drone.

[0180] Among them, the magnetorheological damper pre-installed on the target drone is a device that uses electromagnetic fields to control the flow properties of liquids. By changing the magnitude and direction of the current, the damping force it generates can be adjusted. Therefore, by outputting the control signal, the damping force of the magnetorheological damper can be precisely controlled. In the process of outputting the control signal, the characteristics and response speed of the magnetorheological damper, as well as the dynamic characteristics of the system, need to be considered. By reasonably adjusting the amplitude and frequency of the control signal, the damping force can be precisely controlled and the stability and performance of the system can be ensured. Real-time monitoring and adjustment are required to ensure that the damping force of the magnetorheological damper can effectively reduce the vibration of the drone. This may involve real-time feedback and adjustment of the control signal and the damping force to adapt to the vibration conditions in different flight scenarios and ensure the stability and safety of the system.

[0181] In this way, through the above steps, the vibration of the target drone can be precisely controlled to improve the stability and safety of its flight. For example, suppose that the vibration signal with a vibration frequency of 10Hz is obtained through the acceleration sensor. First, it can be normalized to standardize the data carried by the signal from the range of -10 to 10 to 0 to 1. Then apply a bandpass filter to filter it, retaining the signal in the range of 1-50Hz. Then, after spectrum analysis, the Fourier transform shows that the main frequency component is at 20Hz. Then, through fuzzy control, a control signal is generated. If the vibration frequency is high and the intensity is large, the damping force is increased. And PID control is performed. If there is an error between the actual vibration value and the target value, the damper current is adjusted and set to 0.5A. Among them, the actuator can be used to adjust the magnetorheological damper: the control signal sets the current to 0.5A, changes the magnetic field strength inside the damper, thereby changing the fluid viscosity and reducing vibration.

[0182] Specifically, an optional implementation method is that the implementation process of "adjusting the damping force of the magnetorheological damper pre-set on the target UAV according to the PID control signal" in this step S104 may include: first converting the PID control signal into the control signal required by the magnetorheological damper pre-set on the target UAV, and then adjusting the current or hydraulic pressure input to the magnetorheological damper according to the control signal, changing the magnetorheological properties of the fluid in the magnetorheological damper, so as to adjust the damping force generated by the magnetorheological damper.

[0183] In this implementation, signal conversion is a prerequisite for current regulation. Signal conversion converts the output signal of the PID controller into a current signal, and the current regulation circuit controls the damping force of the magnetorheological damper according to the converted current signal. The specific implementation process and examples are as follows: 1. Signal conversion: Design circuit: Design a signal conversion circuit to convert the voltage output signal of the PID controller (such as 0-10V) into the current signal required by the magnetorheological damper (such as 0-2A). This circuit may include components such as operational amplifiers and voltage-to-current converters. For example: If the output voltage of the PID controller is 5V, the signal conversion circuit converts it into a 1A current signal. The design can be a simple voltage-to-current converter, such as using an operational amplifier and a resistor. 2. Current regulation: Current control circuit: The current regulation circuit receives the current signal after signal conversion and controls the actual current output to the magnetorheological damper. For example, an adjustable current source or a current control module is used to achieve precise current adjustment. For example: If the signal conversion circuit outputs a 1A current signal, the current regulation circuit stably sends this current to the magnetorheological damper to adjust its damping force. If the damping force needs to be adjusted, the output current can be changed through the current regulation circuit, for example, the current can be adjusted from 1A to 1.5A to increase the damping force. In this way, through these two steps, the output signal of the PID controller is effectively converted and adjusted, thereby controlling the damping force of the magnetorheological damper and achieving effective control of the vibration of the drone.

[0184] In the process of generating the control signal, the pre-constructed mathematical model between the damping force and input current of the magnetorheological damper can be used to convert the PID control signal into the target input current of the magnetorheological damper pre-set on the target UAV, and then the input current of the magnetorheological damper is adjusted according to the target input current using the current control circuit as the control signal required by the magnetorheological damper.

[0185] The process of establishing the mathematical model includes the following:

[0186] 1. Data collection: Through experiments or data provided by the manufacturer, collect the damping force value of the magnetorheological damper under different input currents. For example, suppose the collected data is: when the input current is 1A, the damping force is 500N; when the input current is 2A, the damping force is 1000N.

[0187] 2. Model selection: Select a suitable mathematical model to describe the relationship between input current and damping force. Common models include linear models, nonlinear models or more complex fitting functions. For example, select the linear model F = k·I, where F represents the damping force, I represents the input current, and k represents the proportionality coefficient.

[0188] 3. Parameter estimation: Use the collected data to estimate the parameters in the model. Linear regression or other parameter estimation methods can be used to determine the coefficients of the model. For example, through linear regression analysis, the proportionality coefficient k = 500N / A is obtained.

[0189] 4. Model verification: Verify the accuracy of the mathematical model and verify whether the model's prediction results are consistent with the actual damping force through experimental data. If not, the model or parameters need to be adjusted. For example, use different current values ​​for testing to ensure that the damping force predicted by the model matches the actual measured damping force.

[0190] 5. Relationship between mathematical model and control signal:

[0191] Control Signal Generation: A mathematical model is used to convert the current signal output by the PID controller into a damping force signal. The control signal generation process uses this model to determine the required current value to achieve the desired damping force.

[0192] Precise control: By establishing an accurate mathematical model, it can be ensured that the control signal (current value) generated by the PID controller can accurately adjust the damping force of the magnetorheological damper, thereby effectively controlling the vibration of the drone. The mathematical model plays a bridging role in the control signal generation process, converting the signal output by the PID controller into the actual current value that can control the damper, thereby achieving effective control of the drone vibration.

[0193] Combined with a high-frequency control signal generation method, it can ensure that the magnetorheological damper can respond to the output of the PID controller in real time. This real-time control involves the following processing:

[0194] Processing object: Process the output signal of the PID controller. The control signal (such as current value) generated by the PID controller is used to adjust the damping force of the magnetorheological damper.

[0195] High-frequency control: The control signal is generated at a high frequency to ensure that the system can respond quickly to vibration changes. For example, the control signal generation frequency may reach hundreds of hertz to cope with high-frequency vibration.

[0196] Processing results:

[0197] Signal processing: The output signal of the PID controller (such as current value) is converted into a continuous, adjustable current through a high-frequency control signal generation method.

[0198] Actual effect: The MR damper adjusts its damping force in real time according to these high-frequency control signals, responding quickly to vibration changes, thereby reducing the vibration of the drone and improving flight stability. The main task of the real-time control step is to ensure that the control signal output by the PID controller can be processed and adjusted at a high frequency, so that the MR damper can respond quickly and adapt to the vibration of the drone.

[0199] After the control signal is generated, it can also be processed to ensure that it is suitable for the input requirements of the magnetorheological damper. For example, if the control signal voltage or current is insufficient, it needs to be amplified to a level suitable for the magnetorheological damper through a signal amplifier, or the control signal needs to be filtered to remove noise and ensure the purity of the signal. The specific process of signal amplification and filtering in control signal processing is as follows:

[0200] 1. Signal amplification: Assume that the control signal output by the PID controller is 0-1V, but the magnetorheological damper requires an input signal of 0-5V. The amplification process is:

[0201] Signal input: Input the 0-1V signal output by the PID controller to the amplifier.

[0202] Amplifier setting: Use an amplifier with amplification factor (such as a 5x amplifier) ​​to amplify the input signal.

[0203] Output signal: After amplification, the output signal range is 0-5V, which is suitable for the input requirements of magnetorheological damper.

[0204] Results: The adjusted signal amplitude meets the control requirements of the magnetorheological damper and improves the control accuracy and effect.

[0205] 2. Filtering: Assume that the amplified signal still has high-frequency noise in the range of 0-5V. The filtering process is:

[0206] Signal input: Input the amplified signal to the filter.

[0207] Filter selection: Use a low-pass filter to remove high-frequency noise (set the cutoff frequency to 100Hz, for example).

[0208] Filtering operation: The filter outputs a smooth signal by removing frequency components higher than 100Hz.

[0209] Output signal: The signal after noise removal is still 0-5V, but does not contain high-frequency interference.

[0210] Result: The filtered signal is clean and stable, and is more suitable for controlling the magnetorheological damper to achieve more precise vibration control.

[0211] Another implementation method is that the design of the preset actuator interface may include but is not limited to electrical interfaces and mechanical interfaces. Among them, the electrical interface is to design and implement an electrical interface that is adapted to the magnetorheological damper to ensure stable and reliable signal transmission. Its design purpose is to ensure that the control signal can be stably and reliably transmitted to the magnetorheological damper to achieve precise control. Connect the control signal generation system (such as PID controller and signal conversion circuit) to ensure the transmission of the control signal to the magnetorheological damper. It is used in situations where it is necessary to ensure accurate transmission and conversion of control signals, such as current regulation and control signal transmission. It usually involves electrical connection design, including cables, connectors, signal conversion circuits, etc. The mechanical interface is to design and implement the mechanical interface of the magnetorheological damper to ensure that it can be effectively installed on the drone and closely integrated with the body structure to ensure that it will not fall off or loosen during flight and play the best vibration reduction effect. It is mostly used to ensure that the equipment is installed stably, such as installing a magnetorheological damper in a suitable position on the drone. It involves mechanical design and fixing devices, such as brackets, screws, mounting holes, etc. The mechanical interface and the electrical interface work together to enable the magnetorheological damper to accurately respond to the control signal and provide a stable damping force to control the vibration of the drone.

[0212] In addition, to improve the control effect, sensors can be installed near the magnetorheological damper to monitor its actual damping force, temperature and other parameters in real time, and feed the monitoring data back to the control system to dynamically adjust the control signal to ensure precise control of the damping force.

[0213] The control system refers to the entire system used to adjust and manage the magnetorheological damper, which specifically includes the following mechanisms:

[0214] Vibration sensor: used to monitor the vibration of the drone in real time and feed the data back to the control system.

[0215] PID controller: Calculates the control signal based on the feedback data and adjusts the output signal through the proportional, integral and differential parts.

[0216] Filter: Remove noise from the signal to ensure signal quality.

[0217] Amplifier: adjusts the signal amplitude to meet the requirements of the actuator.

[0218] Magnetorheological damper: adjusts the damping force according to the control signal to control the vibration of the drone.

[0219] Data acquisition system: collects vibration data in real time and feeds it back to the controller.

[0220] Feedback adjustment module: adjusts the controller output according to the data fed back by the sensor and optimizes the damping force of the damper.

[0221] The specific process is as follows:

[0222] Data acquisition: The vibration sensor transmits the vibration data of the drone to the control system in real time.

[0223] Data processing: The controller (such as PID controller) analyzes the feedback data and calculates the required control signal.

[0224] Signal Conditioning: Signal processing blocks such as filters and amplifiers condition the quality and amplitude of control signals.

[0225] Actuation adjustment: The adjusted signal is applied via the actuator (MR damper) to adjust the damping force in real time.

[0226] Feedback: The monitoring results are returned to the control system to dynamically adjust the control signal to ensure the accuracy of vibration control.

[0227] The feedback adjustment process involves the following steps:

[0228] Real-time monitoring data: The vibration sensor monitors the actual vibration data of the drone and sends it to the control system.

[0229] Calculated error: The control system calculates the error between the actual vibration data and the desired target. For example, if the target vibration level is 0.5m / s 2 , while the actual measurement is 0.7m / s 2 , then the error is 0.2m / s 2 .

[0230] Update control signal:

[0231] PID controller adjustment: The PID controller uses the error value to adjust the control signal. Assuming the current PID output is 2V, an increase in the error may cause the PID adjustment to increase the output voltage to 2.5V.

[0232] Signal processing: The processed control signal is amplified by an amplifier to ensure that it can drive the magnetorheological damper. For example, the PID output is adjusted from 2.5V to 3V to provide sufficient damping force.

[0233] Adjust the actuator: According to the new control signal (such as 3V), the damping force of the magnetorheological damper increases, thereby reducing vibration. The adjustment process brings the damping force closer to the desired level.

[0234] Feedback and loops:

[0235] Monitoring feedback: The vibration sensor again measures the new vibration level and feeds it back to the control system.

[0236] Dynamic Adjustment: The system continues to adjust the control signal based on new data to ensure that the vibration control accuracy remains within the ideal range.

[0237] Example: The vibration sensor measures a vibration level of 0.8m / s 2 The expected value is 0.5m / s 2 , the error is 0.3m / s 2 The PID controller increases the control signal output to 3.2V. The amplifier adjusts the signal to ensure that 3.2V is output to the MR damper. The actuator is adjusted, such as the MR damper, to adjust the damping force according to 3.2V. The feedback process is: the sensor measures the vibration again and feeds the data back to the control system for the next round of adjustments.

[0238] Before actual application, the entire vibration control system can be fully tested through simulation software to verify the control effect and stability. In addition, field tests can be carried out on drones to verify the actual effect of the control system and make necessary adjustments and optimizations. The specific implementation process is as follows:

[0239] 1. Simulation test

[0240] Objective: To verify the effectiveness and stability of the control system in both theoretical and simulated environments.

[0241] Steps: Build a model. Build a mathematical model of the UAV vibration control system in the simulation software, including PID controller, magnetorheological damper, sensor, etc. Set parameters. Enter the parameters of the control system, such as PID parameters, damper characteristics, etc. Run the simulation and input various simulated vibration signals (such as periodic vibration, random disturbance). Observe the response of the system, such as the effect of vibration reduction and system stability. Analyze the results and check indicators such as system response time, steady-state error and overshoot. Confirm whether the system can effectively control vibration under different vibration conditions. Adjust and optimize. Adjust the PID parameters and other system settings according to the simulation results. Re-run the simulation to verify the optimization effect.

[0242] 2. Experimental Verification

[0243] Objective: To verify the effect of simulation results in the actual UAV and make practical adjustments.

[0244] Steps: Install the system, install the control system on the actual drone, including sensors, PID controller, magnetorheological damper, etc. Conduct field tests, test under different flight conditions (such as static test, dynamic flight test). Record actual vibration data and control signals. Analyze the measured results, compare the actual test data with the simulation results, and analyze the performance of the system in the real environment. Check the stability and actual effect of the control system, such as whether the vibration is effectively controlled. Adjust and optimize, adjust the control parameters and system settings according to the actual test data. Conduct multiple tests to ensure that the optimized system can work stably.

[0245] For example:

[0246] 1. Simulation test:

[0247] Modeling: Build the UAV model in MATLAB / Simulink, including the controller and damper.

[0248] Simulation: Set up different disturbance inputs, such as rapidly changing vibrations.

[0249] Verify: Run the simulation to check how the PID control output adjusts the damping force and observe whether the vibration response is as expected.

[0250] 2. Experimental verification:

[0251] Installation: Install vibration sensors and magnetorheological dampers on the drone.

[0252] Testing: Conduct flight tests to simulate various flight conditions.

[0253] Adjustment: If the vibration control effect is found to be not as good as expected in the actual test, adjust the PID parameters and test again.

[0254] Through these two steps, it is ensured that the control system is not only effective in theory, but also can stably control the vibration of the drone in practical applications.

[0255] In the above steps, another implementation method is to use a precise current control circuit and a high-resolution signal converter to ensure high-precision correspondence between the control signal and the damping force, thereby improving the accuracy of vibration control, reducing the vibration of the drone, and improving flight stability.

[0256] Among them, the role of the precise current control circuit is to control the current to adjust the damping force of the magnetorheological damper. Ensure that the current change is accurately transmitted to the damper, thereby controlling its damping characteristics. In the implementation process, high-precision current sources and regulation circuits are usually involved, such as using high-precision current amplifiers and current feedback control circuits. These circuits can provide stable current output and accurately control the damping force of the damper. The role of the high-resolution signal converter is to convert analog signals (such as the output signal of the PID controller) into digital signals and accurately map them to the control circuit. This ensures a high-precision correspondence between the control signal and the damping force. In the implementation process, a high-resolution analog-to-digital converter (ADC) can be used to convert analog signals into high-precision digital signals to ensure that the control signal is accurately processed and transmitted in the digital control circuit.

[0257] For example: The design process of the current control circuit is: design the current control loop, including the current amplifier and feedback control circuit to ensure the current regulation accuracy. The implementation process is: use the current source design such as the operational amplifier to ensure the stable output of the current and adjust the current according to the control signal. The design process of the signal converter is: select a high-resolution analog-to-digital converter (such as a 24-bit ADC) to ensure the accurate conversion of the analog signal. The implementation process is: connect the analog output of the PID controller to the ADC, and the ADC converts the analog signal into a digital signal for further digital processing and control.

[0258] The combination of these two circuits ensures high accuracy and stability in the transmission of control signals from the PID controller to the MR damper, ultimately achieving effective vibration control.

[0259] It should be noted that in the above steps, an adaptive control algorithm is introduced to dynamically adjust the control parameters according to real-time monitoring data to optimize the damping force control effect. This can improve the system's adaptability and robustness, and maintain good vibration control effects in different flight environments. The adaptive control algorithm adjustment process is as follows:

[0260] 1. Real-time data monitoring:

[0261] Steps: Collect the vibration data and other related information (such as acceleration and vibration frequency) of the drone in real time.

[0262] Tools: Use sensors and data acquisition systems.

[0263] 2. Data analysis:

[0264] Steps: Analyze real-time data to identify vibration patterns and their changes.

[0265] Tools: Apply data processing and analysis algorithms (e.g., time domain analysis, spectral analysis).

[0266] 3. Adaptive algorithm application:

[0267] Steps: Adjust control parameters based on analysis results. Commonly used adaptive control algorithms include model reference adaptive control (MRAC) and adaptive gain adjustment.

[0268] For example:

[0269] 1. Model reference adaptive control: Compare the performance of the actual system with the expected performance of the reference model, and dynamically adjust the controller parameters to make the actual system output meet the requirements of the reference model.

[0270] 2. Adaptive gain adjustment: Real-time adjustment of P, I, and D gains of PID controller to optimize control effect based on real-time feedback.

[0271] 3. Parameter adjustment:

[0272] Steps: Adjust the parameters (such as proportional gain, integral gain, and derivative gain) of the PID controller or other control system based on real-time data and algorithm output.

[0273] Tool: Update the controller's parameter settings and implement them via a feedback loop.

[0274] 4. Control signal generation:

[0275] Step: Apply the adjusted control parameters to control signal generation, affecting the setting of the damping force.

[0276] Tool: Update current regulation circuits or signal processing modules in control systems.

[0277] Specific example:

[0278] Assuming that real-time monitoring data shows that the drone has large vibrations within a certain frequency range, the adaptive control algorithm can be used to make the following adjustments:

[0279] Data analysis: The peak vibration frequency was identified as 50 Hz.

[0280] Algorithm adjustments:

[0281] Model Reference Adaptive Control: Based on the changes in vibration frequency and amplitude, the damping force of the damper is adjusted to provide higher damping at a frequency of 50Hz.

[0282] Adaptive gain adjustment: Increase the integral gain of the PID controller to reduce the steady-state error and vibration amplitude of the system.

[0283] Through these steps, the adjustment of damping force will be reflected in the control system of the UAV in real time, thereby optimizing flight stability and control effect.

[0284] It should also be noted that in the above steps, the data of multiple sensors (such as acceleration, temperature, and pressure) are combined to comprehensively analyze and adjust the control signal to ensure that the damper can work normally under various conditions. This improves the reliability and adaptability of the system and ensures that effective vibration control can be achieved in various complex environments. The specific implementation process is as follows:

[0285] 1. Data collection:

[0286] Steps: Collect real-time data from multiple sensors (such as accelerometer, temperature sensor, pressure sensor).

[0287] Tools: Data acquisition system, which converts sensor signals into digital data and stores them.

[0288] 2. Data Fusion:

[0289] Step: Combine sensor data from different sources into a comprehensive dataset.

[0290] Method: Data fusion algorithm: such as Kalman Filter or weighted average method.

[0291] Example: Combine acceleration data with temperature data to determine whether the acceleration change is abnormal at high temperatures.

[0292] 3. Data analysis:

[0293] Step: Analyze the fused data to identify potential problems or anomalies.

[0294] Methods: Time domain analysis: Check the trend of signal changes. Frequency domain analysis: Use Fourier transform to analyze different frequency components. Statistical analysis: For example, mean and variance calculation.

[0295] 4. Control signal adjustment:

[0296] Steps: Adjust the control signal based on the analysis results to optimize the working state of the damper.

[0297] Method: Fuzzy control: adjust the damping force of the damper according to fuzzy rules. PID adjustment: adjust the parameters of the PID controller according to real-time data.

[0298] 5. Verification and feedback:

[0299] Steps: Apply the adjusted control signal and monitor its effect.

[0300] Method: Real-time monitoring: Track the effect of adjustments in real time through sensors. Feedback mechanism: Feedback the effect data to the control system for dynamic adjustment.

[0301] An example of the specific implementation process:

[0302] Assume that the flight data of the drone is as follows:

[0303] Accelerometer: Detects an increase in vibration amplitude.

[0304] Temperature sensor: A temperature rise was detected.

[0305] Pressure sensor: The pressure remains stable.

[0306] The comprehensive analysis and adjustment process is:

[0307] Data fusion: Use Kalman filter to fuse acceleration, temperature and pressure data to obtain more accurate vibration and environmental status information.

[0308] Data Analysis: The analysis found that rising temperatures can cause the material to expand, which can affect vibration characteristics. Acceleration changes increase under high temperature conditions, indicating that the damper may need to be adjusted to accommodate high temperatures.

[0309] Control signal adjustment: Fuzzy control: Set fuzzy rules, such as "if the temperature is high and the acceleration is large, increase the damping force." PID adjustment: Adjust the integral gain of the PID controller according to temperature changes to adapt to the impact of environmental changes.

[0310] Verification and feedback: Implement the adjusted control signal and observe the changes in vibration and temperature. Adjust the control signal through the feedback mechanism to ensure the best performance of the damper under different conditions.

[0311] This comprehensive analysis and adjustment method can ensure that the damper can effectively control the vibration of the UAV under various operating conditions, improving the stability and reliability of the system.

[0312] It should also be noted that in the above steps, an intelligent fault diagnosis algorithm can also be introduced to monitor the working status of the magnetorheological damper in real time, identify and warn of potential faults. This can improve the safety and maintenance efficiency of the system and reduce the impact of faults on the flight of the drone. Specific implementation steps:

[0313] 1. Data collection:

[0314] Steps: Collect data from the magnetorheological damper and related sensors (such as temperature sensor, current sensor, vibration sensor) in real time.

[0315] Tools: High-frequency data acquisition system to ensure the real-time and accuracy of data.

[0316] 2. Data preprocessing:

[0317] Steps: Preprocess the collected raw data, including denoising, normalization and filtering.

[0318] Methods: Filtering: Use a low-pass or band-pass filter to remove high-frequency noise. Normalization: Adjust the data to a uniform range for subsequent analysis.

[0319] 3. Feature extraction:

[0320] Steps: Extract key features from the preprocessed data, such as vibration amplitude, temperature change rate, current fluctuation, etc.

[0321] Method: Time domain features: such as mean and standard deviation. Frequency domain features: Use Fourier transform to extract the main frequency components.

[0322] 4. Intelligent fault diagnosis algorithm:

[0323] Steps: Apply machine learning or deep learning algorithms to analyze the extracted features and identify potential faults.

[0324] Methods: Support vector machine: Build a classification model to distinguish between normal and faulty states. Neural network: Train a neural network model for fault prediction, learning based on historical data and features. Anomaly detection: Use adaptive thresholds or statistical methods to identify abnormal patterns.

[0325] 5. Fault warning and handling:

[0326] Steps: Perform fault warning and processing based on the output of the intelligent fault diagnosis algorithm.

[0327] Method: Early warning: When a potential fault is detected, an alarm is issued. Processing: Take corresponding measures based on the early warning information, such as adjusting the control strategy or performing maintenance.

[0328] An example of the specific implementation process: Assume that the following data of the magnetorheological damper is monitored:

[0329] Vibration sensor: Detects a sudden increase in vibration amplitude.

[0330] Temperature sensor: The temperature is gradually rising.

[0331] Current sensor: Abnormal current fluctuation.

[0332] The implementation process is:

[0333] Data collection: Get data from the above sensors in real time.

[0334] Data preprocessing: Filtering: Low-pass filter the vibration data to remove high-frequency noise. Normalization: Normalize the vibration amplitude, current, and temperature data to the range of 0-1. Feature extraction: Calculate the mean and standard deviation of the vibration amplitude. Extract the frequency characteristics of current fluctuations. Calculate the temperature change rate.

[0335] Intelligent fault diagnosis algorithm: Training: Use historical data to train the support vector machine (SVM) to identify normal and faulty states. Prediction: Input the extracted features into the trained model to predict whether there is a fault. Anomaly detection: Trigger an early warning when a feature is detected outside the normal range.

[0336] Fault warning and processing: Warning: When the algorithm detects an abnormality, the system will issue an alarm. Processing: Automatically adjust the damper control strategy to reduce vibration; at the same time, notify maintenance personnel to conduct inspections.

[0337] Through this intelligent fault diagnosis system, the status of the magnetorheological damper can be monitored in real time, potential faults can be discovered in time, the risk of failure can be reduced, and the reliability and safety of the system can be improved.

[0338] It should also be noted that in the above steps, the generation and transmission process of the control signal can also be optimized to ensure that the signal is transmitted to the magnetorheological damper with low latency and achieve rapid response. This can significantly improve the real-time performance and response speed of the system and ensure the vibration control effect in a high dynamic environment.

[0339] The specific implementation method of the generation and transmission process of the optimization control signal is not limited, and can be but not limited to the following steps:

[0340] 1. Hardware Optimization

[0341] High frequency signal generation:

[0342] Step 1: Use a high-speed digital-to-analog converter (DAC) to generate control signals to support high-frequency updates.

[0343] For example, choosing a DAC with a high sampling rate (such as 1MHz or higher) ensures that the control signal can be transmitted to the magnetorheological damper at a fast enough speed.

[0344] Efficient signal transmission circuit:

[0345] Steps: Design low-latency, low-noise circuits for signal transmission.

[0346] For example: Use high-bandwidth signal transmission lines and optimized circuit board design to reduce delays and interference during signal transmission.

[0347] 2. Software Optimization

[0348] Real-time control algorithm:

[0349] Steps: Optimize the code of the PID control algorithm and signal generation algorithm to reduce computational delays.

[0350] For example: Use efficient algorithms and programming languages ​​(such as C / C++) and ensure that the code can execute quickly on a real-time operating system (RTOS).

[0351] Interrupt handling:

[0352] Steps: Use priority interrupt mechanism to ensure fast processing of control signals.

[0353] Example: Configuring an interrupt service routine (ISR) to handle emergency control signals to ensure priority response when important events occur.

[0354] 3. System Optimization

[0355] Optimizing feedback loops:

[0356] Steps: Improve the speed of feedback data collection and processing to ensure that the system can respond quickly to control signals.

[0357] Example: Use high-speed data acquisition cards and real-time data processing algorithms to reduce data transmission delays from sensors to control systems.

[0358] Reduce latency:

[0359] Steps: Optimize the interfaces between the various parts of the system to reduce latency.

[0360] For example: In control system design, choose low-latency communication protocols (such as CAN bus) and efficient signal transmission paths.

[0361] 4. Practical application examples

[0362] Optimizing signal generation:

[0363] Current situation: Using a standard DAC, the control signal is updated 10 times per second, resulting in control delays.

[0364] Optimization measures: Replace with a high sampling rate DAC, updating 1000 times per second.

[0365] Effect: The response time of the control signal is shortened from 100ms to 1ms, improving the real-time performance of the system.

[0366] Optimized signal transmission:

[0367] Current Situation: There is high resistance and interference in the signal transmission line, causing delays and signal distortion.

[0368] Optimization measures: Use low-resistance, high-bandwidth transmission lines and shielding materials to reduce interference.

[0369] Effect: Signal transmission delay is reduced, transmission quality is improved, and system response is faster. It can effectively optimize the generation and transmission process of control signals, ensure that the signal is transmitted to the magnetorheological damper with low delay, achieve fast response, and thus improve the stability and control accuracy of the drone.

[0370] In this way, the combination of fuzzy control and PID control can simultaneously use fuzzy control to process the nonlinear and uncertain parts of the vibration signal of the target UAV, and use PID control to provide stable linear control, thereby achieving precise control of the damping force of the magnetorheological damper, reducing the vibration of the target UAV and improving flight stability and safety.

[0371] Next, to facilitate understanding of the vibration control method of the drone provided in the present application, the present application provides an overall example to illustrate the overall implementation process of the vibration control of the drone:

[0372] 1. Signal acquisition and processing. The steps include: signal acquisition, using high-precision acceleration sensors to collect vibration signals at the rotor root, near the engine, and in the center of the fuselage of the drone. Signal preprocessing, converting the analog signal into a digital signal through an analog-to-digital converter, and then applying a low-pass filter to remove high-frequency noise and retain the effective vibration signal. The input part is the original analog vibration signal (for example: acceleration data at the root of the rotor). The output part is a digital vibration signal after filtering (for example: acceleration signal in the range of 0-1V). The optimization methods used include: filter optimization, selecting a suitable low-pass or band-pass filter to remove noise of a specific frequency. And signal amplification, that is, if necessary, adjust the signal amplitude through a signal amplifier to make it suitable for subsequent processing.

[0373] 2. Fuzzy PID control. The steps include: fuzzy controller design, formulating fuzzy rules (such as "if the vibration signal is high and the frequency is low, the control output is small"), and implementing fuzzy logic operations. PID controller debugging, adjusting the proportional (P), integral (I), and differential (D) parameters so that it can effectively control vibration under different conditions. The input part is the vibration signal after signal preprocessing (for example: standardized acceleration data). The output part is the adjustment value generated by the fuzzy controller and the control signal calculated by the PID controller. The optimization methods used include: fuzzy rule optimization, adjusting fuzzy rules through expert experience and simulation. PID parameter optimization, using Ziegler-Nichols method, genetic algorithm, etc. to optimize PID parameters.

[0374] 3. Interface design between PID controller output and magnetorheological damper control signal. The steps include: circuit design, designing electrical interface, converting the signal output by PID controller into a control signal suitable for magnetorheological damper. Signal conversion, using current control circuit to adjust signal amplitude and frequency to match the requirements of damper. The input part is the output signal of PID controller (for example: 0-10V voltage signal). The output part is the adjusted current signal (for example: the current signal for controlling magnetorheological damper, the range may be 0-5A). The optimization methods adopted include: current regulation, ensuring signal stability through precise current control circuit. Signal conversion, designing high-resolution signal converter to improve signal accuracy.

[0375] 4. The control signal is output to the actuator. The steps include: actuator adjustment, outputting the optimized control signal to the magnetorheological damper, adjusting its damping force, and controlling the vibration of the drone. The input part is the final current control signal (for example: the signal adjusted by the current control circuit). The output part is the damping force of the magnetorheological damper (for example: the actual output damping force is adjusted to the required level). The optimization methods used include: real-time control, through the high-frequency signal generation method, to ensure that the magnetorheological damper can respond to the output of the PID controller in real time. Feedback adjustment, dynamically adjust the control signal to ensure precise control of the damping force.

[0376] The summary includes: First, signal acquisition and processing: processing the original vibration signal to ensure data accuracy. Then fuzzy PID control: optimizing vibration control through fuzzy control and PID control. Next, interface design: converting the control signal into a format suitable for the magnetorheological damper. Then control signal output: finally adjusting the magnetorheological damper to achieve vibration control. Each step is subject to detailed optimization to ensure the stability and efficiency of the overall system.

[0377] In summary, the vibration control method of a drone provided in this embodiment first obtains the vibration signal generated by the target drone through a sensor pre-set on the target drone; then analyzes and processes the vibration signal to extract vibration characteristic data, and inputs the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal, and then inputs the fuzzy control signal into a PID controller, processes the fuzzy control signal in three ways, namely, proportional, integral and differential, to generate a PID control signal, and then adjusts the damping force of the magnetorheological damper pre-set on the target drone according to the PID control signal to control and reduce the vibration of the target drone.

[0378] It can be seen that since the present application uses a fuzzy controller to process the complexity and uncertainty of the vibration signal of the target UAV, and combines it with the PID controller to achieve precise control of the damping force of the magnetorheological damper, the vibration of the target UAV can be reduced by adjusting the damping force generated by the magnetorheological damper, thereby achieving more precise, smarter and more economical vibration control of the target UAV, and also improving the flight stability and safety of the target UAV.

[0379] Second embodiment

[0380] This embodiment will introduce a vibration control device for an unmanned aerial vehicle. For related content, please refer to the above method embodiment.

[0381] See also Figure 2 , is a schematic diagram of the composition of a vibration control device for an unmanned aerial vehicle provided in this embodiment, the device comprising:

[0382] An acquisition unit 201 is used to acquire a vibration signal generated by the target UAV through a sensor pre-set on the target UAV;

[0383] An extraction unit 202 is used to analyze and process the vibration signal, extract vibration characteristic data, and input the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal;

[0384] A generating unit 203 is used to input the fuzzy control signal into a proportional-integral-differential (PID) controller, process the fuzzy control signal in three ways: proportional, integral and differential, and generate a PID control signal;

[0385] The control unit 204 is used to adjust the damping force of the magnetorheological damper pre-set on the target UAV according to the PID control signal, so as to control and reduce the vibration of the target UAV.

[0386] In an implementation of this embodiment, the vibration signal is an analog signal; the extraction unit 202 includes:

[0387] A first conversion subunit is used to perform analog-to-digital conversion on the vibration signal to obtain a vibration digital signal corresponding to the vibration signal;

[0388] A filter subunit, used for filtering the vibration digital signal to obtain the filtered vibration signal;

[0389] A normalization subunit, used for performing normalization processing on the filtered vibration signal to obtain the normalized vibration signal;

[0390] The spectrum analysis subunit is used to perform spectrum analysis on the normalized vibration signal to extract vibration characteristic data corresponding to the vibration signal.

[0391] In an implementation of this embodiment, the filtering subunit is specifically used for:

[0392] The vibration digital signal is filtered using a low-pass filter and / or a band-pass filter to obtain the filtered vibration signal.

[0393] In an implementation of this embodiment, the spectrum analysis subunit is specifically used for:

[0394] The normalized vibration signal is subjected to Fourier transform and / or wavelet transform processing to extract vibration characteristic data corresponding to the vibration signal.

[0395] In an implementation of this embodiment, the extracting unit 202 further includes:

[0396] A signal decomposition subunit, used to perform signal decomposition on the vibration characteristic data to obtain frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data;

[0397] The extraction unit 202 further includes:

[0398] A fuzzification subunit, used for inputting the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data into a fuzzy controller for fuzzification processing to obtain fuzzy linguistic variables;

[0399] A first generating subunit is used to perform fuzzy logic operation processing on the fuzzy language variable according to a pre-constructed fuzzy rule to generate an output value of fuzzy control;

[0400] The defuzzification subunit is used to perform defuzzification processing on the output value of the fuzzy control to obtain a fuzzy control signal.

[0401] In an implementation of this embodiment, the generating unit 203 includes:

[0402] A processing subunit, used for normalizing and filtering the fuzzy control signal to obtain a processed fuzzy control signal;

[0403] The second generating subunit is used to input the processed fuzzy control signal into the PID controller, process the linear part of the input signal by three methods of proportion, integration and differentiation, and generate a PID control signal.

[0404] In an implementation of this embodiment, the control unit 204 includes:

[0405] A second conversion subunit, used to convert the PID control signal into a control signal required by the magnetorheological damper pre-set on the target UAV;

[0406] The adjustment subunit is used to adjust the current or hydraulic pressure input to the magnetorheological damper according to the control signal, change the magnetorheological characteristics of the fluid in the magnetorheological damper, and adjust the damping force generated by the magnetorheological damper.

[0407] In an implementation of this embodiment, the second conversion subunit is specifically used for:

[0408] Using a pre-built mathematical model between the damping force of the magnetorheological damper and the input current, the PID control signal is converted into a target input current input to the magnetorheological damper pre-set on the target UAV;

[0409] According to the target input current, the input current of the magnetorheological damper is adjusted by using a current control circuit as a control signal required by the magnetorheological damper.

[0410] In an implementation of this embodiment, the installation position of the sensor pre-set on the target UAV includes at least one of a rotor root position, a position near the engine, and a fuselage center position.

[0411] Furthermore, an embodiment of the present application also provides a vibration control device for a drone, comprising: a processor, a memory, and a system bus;

[0412] The processor and the memory are connected via the system bus;

[0413] The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes any one of the implementation methods of the above-mentioned vibration control method for the drone.

[0414] Furthermore, an embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes any one of the implementation methods of the above-mentioned vibration control method for the drone.

[0415] Furthermore, an embodiment of the present application also provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any one of the implementation methods of the above-mentioned vibration control method for a drone.

[0416] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0417] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0418] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0419] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vibration control method for an unmanned aerial vehicle, characterized in that: include: Acquire a vibration signal generated by the target UAV through a sensor pre-set on the target UAV; Analyzing and processing the vibration signal to extract vibration characteristic data, and inputting the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal; The fuzzy control signal is input into a proportional-integral-differential (PID) controller, and the fuzzy control signal is processed by three methods: proportional, integral and differential to generate a PID control signal; According to the PID control signal, the damping force of the magnetorheological damper pre-set on the target UAV is adjusted to control and reduce the vibration of the target UAV.

2. The method according to claim 1, characterized in that: The vibration signal is an analog signal; the analysis and processing of the vibration signal to extract vibration characteristic data includes: Performing analog-to-digital conversion on the vibration signal to obtain a vibration digital signal corresponding to the vibration signal; Performing filtering processing on the vibration digital signal to obtain the filtered vibration signal; Normalizing the filtered vibration signal to obtain the normalized vibration signal; The normalized vibration signal is subjected to spectrum analysis processing to extract vibration characteristic data corresponding to the vibration signal.

3. The method according to claim 2, characterized in that After performing spectrum analysis on the normalized vibration signal to extract vibration characteristic data corresponding to the vibration signal, the method further includes: Performing signal decomposition on the vibration characteristic data to obtain frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data; The step of inputting the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal comprises: Inputting the frequency components and vibration characteristic values ​​corresponding to the vibration characteristic data into a fuzzy controller for fuzzification processing to obtain fuzzy linguistic variables; According to the pre-constructed fuzzy rules, the fuzzy language variables are processed by fuzzy logic operations to generate output values ​​of fuzzy control; The output value of the fuzzy control is defuzzified to obtain a fuzzy control signal.

4. The method according to claim 1, characterized in that The fuzzy control signal is input into a proportional-integral-differential (PID) controller, and the fuzzy control signal is processed by three methods, namely, proportional, integral and differential, to generate a PID control signal, including: Normalizing and filtering the fuzzy control signal to obtain a processed fuzzy control signal; The processed fuzzy control signal is input into the PID controller, and the linear part of the input signal is processed by three methods: proportional, integral and differential to generate a PID control signal.

5. The method according to claim 1, characterized in that The step of adjusting the damping force of the magnetorheological damper pre-set on the target UAV according to the PID control signal includes: Converting the PID control signal into a control signal required by a magnetorheological damper pre-set on the target UAV; The current or hydraulic pressure input to the magnetorheological damper is adjusted according to the control signal to change the magnetorheological characteristics of the fluid in the magnetorheological damper so as to adjust the damping force generated by the magnetorheological damper.

6. The method according to claim 5, characterized in that The step of converting the PID control signal into a control signal required by the magnetorheological damper pre-set on the target UAV includes: Using a pre-built mathematical model between the damping force of the magnetorheological damper and the input current, the PID control signal is converted into a target input current input to the magnetorheological damper pre-set on the target UAV; According to the target input current, the input current of the magnetorheological damper is adjusted by using a current control circuit as a control signal required by the magnetorheological damper.

7. The method according to any one of claims 1 to 6, characterized in that: The installation position of the sensor pre-set on the target UAV includes at least one of a rotor root position, a position near the engine, and a fuselage center position.

8. A vibration control device for an unmanned aerial vehicle, characterized in that: include: An acquisition unit, configured to acquire a vibration signal generated by the target UAV through a sensor pre-set on the target UAV; An extraction unit is used to analyze and process the vibration signal, extract vibration characteristic data, and input the vibration characteristic data into a fuzzy controller for processing to obtain a fuzzy control signal; A generating unit, used for inputting the fuzzy control signal into a proportional-integral-differential (PID) controller, processing the fuzzy control signal in three ways, namely, proportional, integral and differential, to generate a PID control signal; The control unit is used to adjust the damping force of the magnetorheological damper pre-set on the target UAV according to the PID control signal to control and reduce the vibration of the target UAV.

9. A vibration control device for an unmanned aerial vehicle, characterized in that: include: Processor, memory, system bus; The processor and the memory are connected via the system bus; The memory is used to store one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the method according to any one of claims 1 to 7.

Images