A vibration control method and a vibration control system for unmanned aerial vehicles (UAVs)

Abstract

This invention relates to a vibration control method and a vibration control system for unmanned aerial vehicles (UAVs), belonging to the field of UAV control technology. The method includes: real-time acquisition of dynamic response data of the UAV's payload platform using an inertial measurement module; time-domain and frequency-domain analysis processing of the dynamic response data to construct a state variable characterizing the current vibration state of the payload platform; wherein the state variable includes identified key disturbance frequency bands; adaptively predicting the time window length based on the key disturbance frequency bands; predicting the vibration trend within the time window length based on the state variable; generating an active vibration compensation control command; and actively suppressing vibration of the payload platform based on the active vibration compensation control command. This application adaptively adjusts the prediction time window length based on the key disturbance frequency bands, allowing the window length to dynamically change with the vibration frequency, thus adapting to frequency variations of the UAV under different flight attitudes, wind speed environments, and payload conditions.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a UAV vibration control method and a UAV vibration control system. Background Technology

[0002] With the increasing demands for spatial information accuracy in low-altitude remote sensing, mapping and modeling, power line inspection, and emergency reconnaissance, UAVs equipped with high-precision payloads such as lidar have become the mainstream method for data acquisition. During actual flight, the UAV's airframe is inevitably affected by various factors, including propeller frequency excitation, arm structural resonance, aerodynamic disturbances, and load-coupled vibrations. These high-frequency, random, and superimposed vibrations are directly transmitted to the payload, causing continuous interference to sensors such as the inertial measurement unit (IMU), Global Navigation Satellite System (GNSS) receiver, and lidar. This results in navigation calculation drift, point cloud distortion, and decreased alignment accuracy of multi-source data, ultimately making it difficult for mapping results to meet the needs of high-precision engineering applications.

[0003] To suppress the aforementioned vibrations, various active vibration suppression methods have emerged in the prior art. These methods typically require predicting the vibration state in the near future based on the currently sensed vibration signal and applying reverse compensation in advance accordingly. However, research has revealed that existing active vibration suppression methods use fixed prediction time windows (e.g., 5ms, 10ms), which are difficult to adapt to the real-time changes in vibration frequency during UAV flight (e.g., when the UAV is in different flight states or performing different flight maneuvers). For example, under different flight attitudes, wind speeds, and load conditions, the vibration amplitude and frequency characteristics of the payload platform change constantly. Fixed windows are prone to two extreme problems: when the frequency is low (e.g., in hovering), the vibration period is long, and the fixed window is too short relative to the vibration period, resulting in insufficient predicted phase lead and the compensation command failing to accurately align with the vibration peak, thus limiting the vibration suppression effect; when the frequency is high (e.g., in high-speed flight), the vibration period is extremely short, and the fixed window is too long relative to the vibration period, even exceeding half a period, leading to fuzzy phase prediction, a sharp increase in linear extrapolation error, and the compensation command potentially being opposite to the actual vibration phase, thus amplifying the vibration. Therefore, a fixed window length is difficult to take into account the actual flight conditions with a large range of frequency variations, and the vibration suppression consistency and reliability cannot meet the requirements of high-precision mapping. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, this invention provides a method for controlling the vibration of unmanned aerial vehicles (UAVs) and a vibration control system for UAVs.

[0005] In a first aspect, embodiments of this application provide a vibration control method for an unmanned aerial vehicle (UAV), comprising: real-time acquisition of dynamic response data of a UAV's payload platform based on an inertial measurement module; wherein the payload platform includes: the inertial measurement module, a positioning module, and a lidar; performing time-domain and frequency-domain analysis processing on the dynamic response data to construct a state quantity characterizing the current vibration state of the payload platform; wherein the state quantity includes identified key disturbance frequency bands; adaptively predicting the length of a time window based on the key disturbance frequency bands; predicting the vibration trend within the time window length based on the state quantity; generating an active vibration compensation control command based on the vibration trend within the time window length, the current dynamic response data, and the key disturbance frequency bands; and actively suppressing vibration of the payload platform based on the active vibration compensation control command; wherein, after active vibration suppression, the data from the inertial measurement module, the positioning module, and the lidar within the time window length are fused.

[0006] Optionally, the adaptive prediction of the time window length based on the key disturbance frequency band includes: determining the center frequency of the key disturbance frequency band; and predicting the time window length based on the center frequency of the key disturbance frequency band and a preset scaling factor.

[0007] Optionally, the adaptive prediction of the time window length based on the key perturbation frequency band includes: determining the optimal phase lead angle; determining the center frequency of the key perturbation frequency band; and predicting the time window length based on the center frequency of the key perturbation frequency band and the optimal phase lead angle.

[0008] Optionally, determining the optimal phase lead angle includes: acquiring the response delay time between the actuator receiving the historical vibration active compensation control command and generating actual displacement; calculating the optimal phase lead angle based on the center frequency of the key disturbance frequency band and the response delay time; wherein the optimal phase lead angle is positively correlated with the product of the response delay time and the center frequency of the key disturbance frequency band.

[0009] Optionally, determining the optimal phase lead angle includes: calculating the first derivative of the center frequency of the key disturbance frequency band with respect to time as the frequency change rate; and calculating the optimal phase lead angle based on the frequency change rate, a preset reference phase lead angle, and the center frequency of the key disturbance frequency band.

[0010] Optionally, the step of performing time-domain and frequency-domain analysis on the dynamic response data to construct a state variable characterizing the current vibration state of the load platform includes: constructing a time-domain state model characterizing the discrete state evolution form based on the dynamic response data; performing frequency-domain transformation on the acceleration signal in the dynamic response data to construct a vibration spectrum function; wherein the key disturbance frequency band is determined by the vibration spectrum function; and constructing a state variable characterizing the current vibration state of the load platform based on the time-domain state model and the vibration spectrum function.

[0011] Optionally, determining the key disturbance frequency band includes: based on the vibration spectrum function, extracting the frequency with the largest vibration energy amplitude from the frequency domain distribution characteristics as the key disturbance frequency band.

[0012] Optionally, predicting the vibration trend within the time window length based on the state variables includes: constructing a state prediction function; wherein the state prediction function characterizes the relationship between the load vibration displacement at the current moment, the load vibration velocity at the current moment, the load vibration acceleration at the current moment, and the load vibration displacement within the time window length; and predicting the vibration trend within the time window length based on the state prediction function and the state variables; wherein the vibration trend within the time window length is the load vibration displacement within the time window length.

[0013] Optionally, generating the vibration active compensation control command based on the vibration trend, current dynamic response data, and key disturbance frequency band within the time window length includes: multiplying the load vibration displacement within the time window length, the vibration velocity component in the current dynamic response data, and the vibration energy amplitude of the key disturbance frequency band by corresponding gain coefficients and then performing weighted processing to obtain the vibration active compensation control command.

[0014] Optionally, the active vibration suppression of the load platform based on the active vibration compensation control command includes: decomposing the active vibration compensation control command into low-frequency components and high-frequency components; outputting the low-frequency components to a macro-displacement actuator to perform low-frequency large-stroke displacement compensation on the load platform; and outputting the high-frequency components to a micro-vibration actuator to suppress high-frequency reverse vibration on the load platform.

[0015] Secondly, this application provides a vibration control system for an unmanned aerial vehicle (UAV), comprising: a vibration sensing module for real-time acquisition of dynamic response data of the UAV's payload platform based on an inertial measurement module; wherein the payload platform includes: the inertial measurement module, a positioning module, and a lidar; performing time-domain and frequency-domain analysis processing on the dynamic response data to construct a state quantity characterizing the current vibration state of the payload platform; wherein the state quantity includes identified key disturbance frequency bands; a prediction module for adaptively predicting the length of a time window based on the key disturbance frequency bands; predicting the vibration trend within the time window length based on the state quantity; and an active vibration suppression module for generating an active vibration compensation control command based on the vibration trend within the time window length, the current dynamic response data, and the key disturbance frequency bands; and actively suppressing vibration of the payload platform based on the active vibration compensation control command; wherein, after active vibration suppression, the data from the inertial measurement module, the positioning module, and the lidar within the time window length are fused.

[0016] The beneficial effects of this invention include: The active vibration suppression method proposed in this application correlates the prediction time window with the current key disturbance frequency in real time. Compared with the fixed window approach in existing active vibration suppression methods, this avoids the problems of insufficient phase lead at low frequencies and ambiguous phase prediction at high frequencies. It maintains precise alignment between compensation commands and actual vibrations across the entire frequency band, significantly improving the consistency of vibration suppression effects. Specifically, this application adaptively adjusts the prediction time window length based on the key disturbance frequency band, allowing the window length to dynamically change with the vibration frequency. This adapts to frequency variations of UAVs under different flight attitudes, wind speeds, and load conditions, fundamentally solving the technical deficiency of fixed windows in accommodating a wide frequency range and improving the adaptability and robustness of active vibration suppression. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of a drone vibration control method provided in an embodiment of the present invention. Figure 2 This is a flowchart illustrating the steps of another UAV vibration control method provided in an embodiment of the present invention; Figure 3 This is a block diagram of a drone vibration control system provided in an embodiment of the present invention; Figure 4 This is a module block diagram of a drone provided in an embodiment of the present invention. Detailed Implementation

[0018] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0019] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] The study found that existing active vibration suppression methods use fixed prediction time windows (e.g., 5ms, 10ms), which are difficult to adapt to real-time changes in vibration frequency during UAV flight (e.g., when the UAV is in different flight states or performing different flight maneuvers). For example, under different flight attitudes, wind speeds, and load conditions, the vibration amplitude and frequency characteristics of the payload platform change constantly. Fixed windows are prone to two extreme problems: when the frequency is low (e.g., in hovering), the vibration period is long, and the fixed window is too short relative to the vibration period, resulting in insufficient predicted phase lead and the compensation command failing to accurately align with the vibration peak, thus limiting the vibration suppression effect; when the frequency is high (e.g., in high-speed flight), the vibration period is extremely short, and the fixed window is too long relative to the vibration period, even exceeding half a period, leading to fuzzy phase prediction, a sharp increase in linear extrapolation error, and the compensation command potentially being opposite to the actual vibration phase, amplifying the vibration. Therefore, fixed window lengths cannot adequately accommodate actual flight conditions with large frequency variations, and the consistency and reliability of vibration suppression cannot meet the requirements of high-precision mapping.

[0021] In view of the above problems, this application proposes the following embodiments to solve the above technical problems.

[0022] Please see Figure 1 This application provides a method for controlling the vibration of an unmanned aerial vehicle (UAV), including steps 101 to 106.

[0023] Step 101: Real-time acquisition of dynamic response data of the UAV's payload platform based on the inertial measurement module.

[0024] The payload platform is an integrated UAV platform, which specifically includes an inertial measurement unit (IMU), a positioning module (such as a GNSS receiver), and a lidar.

[0025] The data types of the dynamic response data mentioned above include displacement, velocity, and acceleration.

[0026] In specific applications, such as when drones are performing surveying, power line inspection and emergency reconnaissance, they can acquire dynamic response data of the payload platform collected by the inertial measurement module in real time.

[0027] Step 102: Perform time-domain and frequency-domain analysis on the dynamic response data to construct state variables that characterize the current vibration state of the load platform.

[0028] The state variables include the identified key disturbance frequency bands. That is, at the time domain level, the dynamic response data is transformed into a time-correlated state description. At the frequency domain level, the frequency band with the highest concentration of vibration energy is identified as the key disturbance frequency band. This key disturbance frequency band typically corresponds to the most destructive vibration sources, such as periodic disturbances generated by the rotation of the UAV propeller or resonance of the boom structure.

[0029] Here, through time-frequency analysis, the original dynamic response data is transformed into a comprehensive state quantity containing time-domain state information and key frequency band information in the frequency domain, so that vibration is transformed from traditional interference noise into a quantifiable and describable control input variable.

[0030] Step 103: Adaptively predict the time window length based on the key disturbance frequency band.

[0031] This application is equivalent to dynamically calculating the length of a time window that matches the key disturbance frequency band after it has been identified. This time window length is then used as the prediction node when predicting future vibration trends.

[0032] Step 104: Based on the state variables, predict the vibration trend over the time window length.

[0033] Based on the state variables characterizing the current vibration, the vibration trend at future moments can be predicted by combining the dynamic characteristics of the load platform.

[0034] Based on the current vibration state and the dynamic evolution law of the load platform, the trend of vibration displacement change in the near future can be inferred.

[0035] Step 105: Generate active vibration compensation control commands based on the vibration trend, current dynamic response data and key disturbance frequency bands within the time window.

[0036] By integrating three types of information—future vibration trends, current dynamic response data, and key disturbance frequency bands—the final active vibration compensation control command is generated. This command combines feedforward prediction based on future trends, real-time feedback based on current data, and targeted suppression based on key frequency bands, resulting in a more comprehensive and precise control effect.

[0037] Step 106: Actively suppress vibration of the load platform based on the active vibration compensation control command.

[0038] Since the inertial measurement module, positioning module, and lidar are all integrated on the same payload platform, the active vibration suppression effect acts on all sensors simultaneously, ensuring the consistency and synchronization of multi-source data.

[0039] In this process, after active vibration suppression, data from the inertial measurement module, positioning module, and lidar are fused within a time window. The resulting high-precision pose information can be used for UAV navigation control or mission planning, while the calibrated extrinsic parameters are directly used for geometric correction of the lidar point cloud, generating high-precision mapping results.

[0040] In summary, the UAV vibration control method provided in this application has the following beneficial effects: The active vibration suppression method proposed in this application correlates the prediction time window with the current key disturbance frequency in real time. Compared with the fixed window approach in existing active vibration suppression methods, it avoids the problems of insufficient phase lead at low frequencies and ambiguous phase prediction at high frequencies. It maintains precise alignment between compensation commands and actual vibrations across the entire frequency band, significantly improving the consistency of vibration suppression effects. That is, by adaptively adjusting the prediction time window length based on the key disturbance frequency band, this application allows the window length to dynamically change with the vibration frequency, adapting to frequency changes of the UAV under different flight attitudes, wind speed environments, and load conditions. This fundamentally solves the technical defect that a fixed window cannot cover a wide frequency range, improving the adaptability and robustness of active vibration suppression.

[0041] The following provides a detailed description of the adaptive prediction time window length in the embodiments of this application.

[0042] Optionally, the above steps, based on the key disturbance frequency band, adaptively predict the time window length, which may specifically include: determining the center frequency of the key disturbance frequency band; and predicting the time window length based on the center frequency of the key disturbance frequency band and a preset scaling factor.

[0043] The calculation of the time window length mentioned above can be specifically performed using the following formula: ; in, This indicates the predicted time window length. This represents the preset scaling factor, which can range from 0.05 to 0.25. This indicates the center frequency of the key disturbance band.

[0044] It should be noted that the above method has low computational complexity and is suitable for real-time implementation in embedded flight control computers. Furthermore, by appropriately setting the scaling factor (0.05~0.25), the window length can be kept within a fixed fraction of the vibration period, thereby maintaining a relatively stable phase lead at different frequencies.

[0045] Optionally, the above steps adaptively predict the time window length based on the key perturbation frequency band, including: determining the optimal phase lead angle; determining the center frequency of the key perturbation frequency band; and predicting the time window length based on the center frequency of the key perturbation frequency band and the optimal phase lead angle.

[0046] The calculation of the time window length mentioned above can be specifically performed using the following formula: ; in, This indicates the predicted time window length. Indicates the optimal phase lead angle. This indicates the center frequency of the key disturbance band.

[0047] It should be noted that this method converts the control parameter from time length to phase lead angle. Specifically, by determining or preset the optimal phase lead angle (such as 90°), consistent phase alignment characteristics can be achieved across the entire frequency range, avoiding fluctuations in phase lead caused by frequency changes.

[0048] Optionally, determining the optimal phase lead angle may specifically include: obtaining the response delay time between the actuator receiving the historical vibration active compensation control command and generating the actual displacement; calculating the optimal phase lead angle based on the center frequency of the key disturbance frequency band and the response delay time; wherein the optimal phase lead angle is positively correlated with the product of the response delay time and the center frequency of the key disturbance frequency band.

[0049] The calculation of the optimal phase lead angle can be specifically performed using the following formula: ; in, Indicates the optimal phase lead angle. Indicates the center frequency of the key disturbance frequency band. Indicates the response delay time. This represents the preset phase margin, ranging from 10° to 30°, used to compensate for model errors and uncertainties.

[0050] It should be noted that this method incorporates the physical characteristics of the actuator into the calculation of the adaptive window, achieving real-time compensation for actuator latency. The response latency of UAV actuators (especially macro-displacement motors) varies significantly under different temperatures, loads, or aging conditions. Without considering these variations, a fixed phase lead angle would lead to actual phase alignment deviations. This solution uses online identification or measurement to dynamically adjust the phase lead angle according to the actuator state, always maintaining precise out-of-phase alignment between the compensation signal and vibration, thus overcoming the adverse effects of actuator performance drift on vibration suppression in traditional methods.

[0051] Optionally, determining the optimal phase lead angle may specifically include: calculating the first derivative of the center frequency of the key disturbance frequency band with respect to time as the frequency change rate; and calculating the optimal phase lead angle based on the frequency change rate, the preset reference phase lead angle, and the center frequency of the key disturbance frequency band.

[0052] The calculation of the optimal phase lead angle can be specifically performed using the following formula: ; in, Indicates the optimal phase lead angle. Indicates the center frequency of the key disturbance frequency band. This indicates the preset reference phase lead angle. Indicates the rate of change of frequency. This represents the adaptive gain coefficient.

[0053] It should be noted that this solution is specifically designed for UAVs operating under variable speed conditions (such as accelerated climb, rapid maneuvering, and decelerated landing). When the rotational speed changes rapidly, the derivative of the vibration frequency is large, and assuming a constant frequency will result in significant prediction errors. Therefore, the above method introduces the rate of frequency change as a correction term, enabling the phase lead angle to predict the frequency change trend: automatically increasing the lead angle when the frequency increases and automatically decreasing the lead angle when the frequency decreases, thereby compensating for the additional phase drift caused by frequency changes.

[0054] Please see Figure 2 Optionally, the above steps perform time-domain and frequency-domain analysis on the dynamic response data to construct state variables characterizing the current vibration state of the load platform, including steps 201 to 203.

[0055] Step 201: Based on dynamic response data, construct a time-domain state model that represents the evolution of the discretized state.

[0056] In this embodiment, the perturbation vibration of the UAV payload during flight is transformed from being treated merely as noise interference into an observable, quantifiable, predictable, and ultimately active control dynamic variable. This transforms vibration from an object of passive isolation and post-processing filtering into an effective input signal in the closed-loop control link.

[0057] First, by using the dynamic response data continuously collected by the accelerometer and gyroscope sampling period in the inertial measurement unit, a time-domain state model is constructed, incorporating displacement, velocity, and acceleration into the same dynamic system for evolution inference, so that the time-domain process of vibration becomes a reconstructable state sequence.

[0058] The state vector Defined as: (1) in, It represents a small displacement of the load in a certain direction; This represents the first derivative of displacement with respect to time, i.e., the vibration velocity; This represents the vibration acceleration observation output in real time by the accelerometer.

[0059] To enable prediction capabilities, a dynamic model is constructed using a discretized state evolution form, namely, the time-domain state model: (2) And set the corresponding observation equations: (3) The IMU sampling period For discrete sampling times, formula (2) describes the time from... Time's up The state transition relationship at any given moment, The angular acceleration input is derived from the gyroscope. and These represent system process noise and measurement noise, respectively. The actual observation output of the sensor. By using formulas (2) and (3), the vibration becomes an estimable and predictable dynamic variable, providing a real-time available state prior for subsequent active vibration suppression, fundamentally breaking through the bottleneck that traditional UAV load vibration can only be compensated in post-processing.

[0060] Step 202: Perform frequency domain transformation on the acceleration signal in the dynamic response data to construct the vibration spectrum function.

[0061] Among them, the key disturbance frequency band is determined by the vibration spectrum function.

[0062] However, time-domain modeling alone is insufficient to effectively identify the resonance caused by the coupling between the periodic disturbances generated by the UAV propeller rotation and the natural frequency of the arm. Therefore, this module further introduces a frequency-domain analysis mechanism, using a windowed fast Fourier transform to construct a vibration spectrum function from the acceleration signal: (4) In the formula, Indicates the vibration signal at angular frequency The energy amplitude at that location, The sampling window length, For the first Acceleration data at any given moment.

[0063] Optionally, determining the key disturbance frequency band includes: based on the vibration spectrum function, extracting the frequency with the largest vibration energy amplitude from the frequency domain distribution characteristics as the key disturbance frequency band.

[0064] Through the Finding the maximum energy component yields the frequency band that has the greatest impact on system disturbance and needs to be suppressed, which can be considered the key disturbance frequency band. : (5) Step 203: Based on the time-domain state model and vibration spectrum function, construct the state variables that characterize the current vibration state of the load platform.

[0065] Finally, the time-domain vibration state (derived from the time-domain state model) and frequency-domain energy information (derived from the vibration spectrum function) are fused to generate state variables that can be used as inputs for active vibration suppression control. : (6) in, , , , , which are weighting coefficients used to adaptively adjust the proportion of the four components in active control according to the flight state of the UAV; the first three terms in formula (6) describe the time-domain characteristics of instantaneous vibration; These four elements directly characterize the most destructive resonant frequency band energy and together constitute a closed-loop control input that can directly act on the vibration damping actuator.

[0066] In summary, by constructing a discretized state evolution model, vibration is transformed into a recursive dynamic system. Key disturbance frequency bands are accurately located through frequency domain transformation, and time-domain state and frequency-domain energy are integrated into a comprehensive state quantity. This transforms the original vibration signal into a control input variable containing rich time-frequency information, providing an accurate and comprehensive data foundation for subsequent vibration trend prediction and targeted active vibration suppression. This fundamentally solves the limitation of traditional methods where vibration is only passively treated as noise.

[0067] Optionally, based on state variables, predicting the vibration trend within a time window includes: constructing a state prediction function; wherein the state prediction function characterizes the relationship between the load vibration displacement at the current moment, the load vibration velocity at the current moment, the load vibration acceleration at the current moment, and the load vibration displacement within the time window; and predicting the vibration trend within the time window based on the state prediction function and state variables; wherein the vibration trend within the time window is the load vibration displacement within the time window.

[0068] The core of the above steps lies in the parallel construction of feedforward prediction and feedback steady-state control, realizing a dual vibration suppression structure that compensates before vibration occurs and corrects vibration after it occurs. This significantly improves the ability to specifically eliminate propeller frequency harmonics and the inherent resonant frequency band of the structure, thereby reducing GNSS / INS (which includes an inertial measurement module) attitude drift from the root and providing a stable physical basis for the geometric accuracy of laser point clouds.

[0069] To achieve the above objectives, we first base our approach on the state vector. A short-term dynamic predictor that can estimate future vibration trends in advance is introduced, and its state prediction function is defined as: (7) Indicates the current time Load vibration displacement, Indicates the current time The load vibration velocity, Indicates the load vibration acceleration at the current moment, Indicates prediction The load vibration displacement after a certain time. , , These are the system state matrix parameters obtained by calibration based on the stiffness, damping characteristics, and inertial parameters of the UAV platform. The value range is the length of the prediction time window and can be adaptively adjusted according to the frequency band to ensure that future estimates do not produce numerical drift.

[0070] By constructing a state prediction function, a quantitative relationship was established between the load vibration displacement, velocity, and acceleration at the current moment and the vibration displacement at future moments. Based on the dynamic characteristics of the load platform, the state prediction function enables the system to infer future changes from the current state, transforming the control mode from traditional post-vibration hysteresis correction to feedforward active control that compensates before vibration occurs.

[0071] After constructing the state prediction function, it is combined with the current vibration state variables obtained in the previous steps to perform the prediction calculation of the vibration trend at future moments.

[0072] Optionally, the above steps generate active vibration compensation control commands based on the vibration trend, current dynamic response data, and key disturbance frequency bands within the time window length. This includes multiplying the load vibration displacement within the time window length, the vibration velocity component in the current dynamic response data, and the vibration energy amplitude of the key disturbance frequency band by the corresponding gain coefficients and then performing weighted processing to obtain the active vibration compensation control commands.

[0073] The prediction results are converted into execution signals, and a composite control law that simultaneously considers frequency domain targeted suppression and time domain trend compensation is proposed, as shown in formula (8): (8) in, The active compensation output of the final drive actuator (corresponding to the final output vibration active compensation control command). , , These are the gain coefficients corresponding to the feedforward position compensation, damped velocity suppression, and frequency domain vibration suppression weighting factors, respectively. Among them, increasing... It can enhance the feedforward compensation effect, making the system more proactive in responding to future trends; increase This can enhance the system's real-time damping capability, enabling the system to quickly quell current disturbances. Increase It can enhance the suppression effect on key resonant frequency bands and make the vibration suppression energy more concentrated. This indicates the key resonance frequency extracted in the previous module. The vibration amplitude envelope at that location.

[0074] This control law introduces future state prediction. This shifts vibration suppression from reactive to feedforward active control, significantly reducing control delay; through Vibration suppression has evolved from full-frequency control to targeted resonant frequency band elimination, improving the suppression efficiency of high-frequency propeller frequencies and structural vibration points.

[0075] By multiplying future vibration displacement (feedforward information), current vibration velocity (real-time feedback information), and energy of key disturbance frequency bands (targeted suppression information) by their respective gain coefficients and then weighting and synthesizing them, a composite control command that takes into account all three functions is generated. This composite control strategy enables the vibration suppression system to respond in advance to upcoming vibrations, suppress current disturbances in real time, and concentrate energy to eliminate the most harmful resonant components, thus achieving synergistic optimization of multi-dimensional control objectives.

[0076] Optionally, the load platform is actively suppressed based on the vibration active compensation control command, including: decomposing the vibration active compensation control command into low-frequency components and high-frequency components; outputting the low-frequency components to the macro-displacement actuator to perform low-frequency large-stroke displacement compensation on the load platform; and outputting the high-frequency components to the micro-vibration actuator to suppress high-frequency reverse vibration on the load platform.

[0077] Low-frequency components typically correspond to vibrations with lower frequencies and larger amplitudes (such as low-frequency displacements caused by aircraft swaying or wind disturbances). High-frequency components typically correspond to vibrations with higher frequencies and smaller amplitudes (such as high-frequency micro-vibrations caused by propeller harmonics or structural resonances).

[0078] In this embodiment, the macro displacement actuator can be a motor-driven displacement compensator; while the micro vibration actuator can be a piezoelectric ceramic actuator.

[0079] In terms of execution, this module not only outputs compensation commands, but also adopts a collaborative mechanism of macro-drive and micro-drive for vibrations in different frequency bands: the macro motor is responsible for low-frequency large-stroke compensation, and the piezoelectric micro-actuator is responsible for high-frequency fine anti-phase drive, so that the platform forms a physical superposition structure of low-frequency rigid support and high-frequency flexible cancellation.

[0080] Please see Figure 3 Based on the same inventive concept, this application also provides a drone vibration control system 30, comprising: The vibration sensing module 301 is used to collect dynamic response data of the UAV's payload platform in real time based on the inertial measurement module; wherein the payload platform includes: the inertial measurement module, the positioning module, and the lidar; the dynamic response data is analyzed and processed in the time domain and frequency domain to construct a state quantity characterizing the current vibration state of the payload platform; wherein the state quantity includes the identified key disturbance frequency bands.

[0081] The prediction module 302 is used to adaptively predict the time window length based on the key disturbance frequency band; and to predict the vibration trend within the time window length based on the state quantity.

[0082] The active vibration suppression module 303 is used to generate an active vibration compensation control command based on the vibration trend, current dynamic response data and key disturbance frequency band within the time window length; and to actively suppress vibration of the load platform based on the active vibration compensation control command; wherein, after active vibration suppression, the data from the inertial measurement module, the positioning module and the lidar within the time window length are fused.

[0083] Please see Figure 4 Based on the same inventive concept, this application also provides a drone 40.

[0084] The drone 40 includes at least: a processing module 401 and a payload platform 402 connected to the processing module 401.

[0085] The payload platform 402 includes: an inertial measurement module 4021, a positioning module 4022, and a lidar 4023.

[0086] The processing module 401 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0087] It should be noted that the above-mentioned systems, drones, etc., are based on the same concept as the method embodiments of this application. The modules designed in the system, as well as the steps performed by the device and the resulting technical effects, can all be found in the method embodiments section, and will not be repeated here.

[0088] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0089] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0090] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0091] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A vibration control method for unmanned aerial vehicles (UAVs), characterized in that, include: The inertial measurement module is used to collect dynamic response data of the UAV's payload platform in real time; wherein, the payload platform includes: the inertial measurement module, the positioning module, and the lidar; The dynamic response data is analyzed and processed in the time domain and frequency domain to construct a state variable characterizing the current vibration state of the load platform; wherein the state variable includes the identified key disturbance frequency bands; Based on the key disturbance frequency band, the adaptive prediction time window length is determined. Based on the state quantity, predict the vibration trend within the time window length; Based on the vibration trend, current dynamic response data and key disturbance frequency band within the time window, a vibration active compensation control command is generated. The load platform is actively vibration suppressed based on the vibration active compensation control command; wherein, after active vibration suppression, the data from the inertial measurement module, the positioning module, and the lidar within the time window are fused.

2. The UAV vibration control method according to claim 1, characterized in that, The adaptive prediction time window length based on the key perturbation frequency band includes: Determine the center frequency of the key disturbance frequency band; The length of the time window is predicted based on the center frequency of the key disturbance frequency band and a preset scaling factor.

3. The UAV vibration control method according to claim 1, characterized in that, The adaptive prediction time window length based on the key perturbation frequency band includes: Determine the optimal phase lead angle; Determine the center frequency of the key disturbance frequency band; The length of the time window is predicted based on the center frequency of the key perturbation band and the optimal phase lead angle.

4. The UAV vibration control method according to claim 3, characterized in that, Determining the optimal phase lead angle includes: The response delay time between the actuator receiving the historical vibration active compensation control command and the actual displacement is obtained; The optimal phase lead angle is calculated based on the center frequency of the key disturbance band and the response delay time. The optimal phase lead angle is positively correlated with the product of the response delay time and the center frequency of the key disturbance band.

5. The UAV vibration control method according to claim 3, characterized in that, Determining the optimal phase lead angle includes: Calculate the first derivative of the center frequency of the key disturbance frequency band with respect to time, and use it as the frequency change rate; The optimal phase lead angle is calculated based on the frequency change rate, the preset reference phase lead angle, and the center frequency of the key disturbance frequency band.

6. The UAV vibration control method according to claim 1, characterized in that, The step of performing time-domain and frequency-domain analysis on the dynamic response data to construct state variables characterizing the current vibration state of the load platform includes: Based on the dynamic response data, a time-domain state model representing the evolution of the discrete state is constructed. The acceleration signal in the dynamic response data is subjected to frequency domain transformation to construct a vibration spectrum function; wherein, the key disturbance frequency band is determined by the vibration spectrum function. Based on the time-domain state model and the vibration spectrum function, a state variable characterizing the current vibration state of the load platform is constructed.

7. The UAV vibration control method according to claim 6, characterized in that, The prediction of the vibration trend within the time window length based on the state quantity includes: Construct a state prediction function; wherein the state prediction function characterizes the relationship between the load vibration displacement at the current moment, the load vibration velocity at the current moment, the load vibration acceleration at the current moment, and the load vibration displacement within the time window length; Based on the state prediction function and the state quantity, the vibration trend under the time window length is predicted; wherein, the vibration trend under the time window length is the load vibration displacement under the time window length.

8. The UAV vibration control method according to claim 7, characterized in that, The generation of active vibration compensation control commands based on the vibration trend, current dynamic response data, and key disturbance frequency bands within the time window includes: The load vibration displacement, the vibration velocity component in the current dynamic response data, and the vibration energy amplitude of the key disturbance frequency band under the specified time window length are multiplied by their respective gain coefficients and then weighted to obtain the vibration active compensation control command.

9. The UAV vibration control method according to claim 1, characterized in that, The active vibration suppression of the load platform based on the active vibration compensation control command includes: The vibration active compensation control command is decomposed into low-frequency components and high-frequency components; The low-frequency component is output to the macro-displacement actuator to perform low-frequency large-stroke displacement compensation on the load platform; The high-frequency component is output to the micro-vibration actuator to suppress high-frequency reverse vibration of the load platform.

10. A vibration control system for unmanned aerial vehicles (UAVs), characterized in that, include: A vibration sensing module is used to collect dynamic response data of the UAV's payload platform in real time based on the inertial measurement module; wherein the payload platform includes: the inertial measurement module, the positioning module, and the lidar; the dynamic response data is analyzed and processed in the time domain and frequency domain to construct a state quantity characterizing the current vibration state of the payload platform; wherein the state quantity includes the identified key disturbance frequency bands; The prediction module is used to adaptively predict the time window length based on the key disturbance frequency band; and to predict the vibration trend within the time window length based on the state quantity. An active vibration suppression module is used to generate an active vibration compensation control command based on the vibration trend, current dynamic response data, and key disturbance frequency band within the time window length; and to actively suppress vibration of the load platform based on the active vibration compensation control command; wherein, after active vibration suppression, the data from the inertial measurement module, the positioning module, and the lidar within the time window length are fused.

Images