Unmanned aerial vehicle hoisting anti-swing control method and system based on variable rope length self-adaptation

Abstract

The application discloses a variable rope length adaptive-based unmanned aerial vehicle hoisting swing control method and system, relates to the unmanned aerial vehicle field, and comprises the following steps: taking real-time rope length as an independent variable, and updating the speed gain and the position gain of swing control in a nonlinear manner; mapping the load position deviation into a position feedback guide component in the speed domain through the updated position gain, mapping the swing state parameter into a swing damping component in the speed domain through the updated speed gain, and generating a Coriolis feedforward compensation component in the speed domain based on the rope length change parameter and the swing state parameter; fusing the position feedback guide component, the swing damping component and the Coriolis feedforward compensation component to obtain a target speed bias vector, and driving the fuselage to generate active following compensation movement based on the target speed bias vector. The application can solve the problem that the anti-swing control fails and flight instability occurs due to nonlinear dynamic disturbance caused by rope length dynamic time variation when the unmanned aerial vehicle performs hoisting operation with a winch.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a UAV hoisting anti-sway control method and system based on variable rope length adaptive control. Background Technology

[0002] A drone-based sling-and-carry system typically consists of the aircraft itself, an onboard servo winch, a flexible sling, and a load at the bottom. During actual sling-and-carry operations, the load at the end inevitably experiences periodic reciprocating oscillations due to the combined effects of multiple sources of disturbance, including aircraft maneuvering, external wind disturbances, and the impact of the winch's retrieval and deployment.

[0003] Existing anti-swing control systems for multi-rotor UAVs are generally designed based on a "fixed-length pendulum" model, with their control gain remaining constant throughout the entire flight cycle. However, in real winch-assisted hoisting operations, the natural frequency of the pendulum system will drift nonlinearly as the rope length is dynamically extended and retracted. This results in control overshoot causing fuselage oscillation when the rope is short, and insufficient damping compensation when the rope is long, especially during the load lowering phase, where the swing amplitude surges dramatically. Summary of the Invention

[0004] This application provides a method and system for anti-sway control of UAV hoisting based on variable rope length adaptation, in order to solve the problem that the traditional "fixed length pendulum" model cannot cope with the nonlinear dynamic disturbance caused by the dynamic time-varying rope length, which leads to the airframe oscillation problem.

[0005] Firstly, this application provides a method for anti-sway control of UAV hoisting based on variable rope length adaptive control, including: To obtain real-time rope length data of the airborne winch in order to calculate rope length variation parameters, and to obtain load swing state parameters; Using the real-time rope length as the independent variable, the velocity gain and position gain of the anti-swing control are updated nonlinearly, and the updated velocity gain and position gain are obtained accordingly. The updated position gain of the load position deviation is mapped to the position feedback guidance component in the velocity domain, the updated velocity gain of the swing state parameter is mapped to the swing damping component in the velocity domain, and the Coriolis feedforward compensation component in the velocity domain is generated based on the rope length change parameter and the swing state parameter. The position feedback guidance component, the swing damping component, and the Coriolis feedforward compensation component are fused to obtain the target velocity offset vector, and the fuselage is driven to generate active following compensation motion based on the target velocity offset vector.

[0006] In an alternative embodiment of the first aspect, the rope length variation parameter includes a rope length variation rate, and the method further includes, before generating the Coriolis feedforward compensation component in the velocity domain based on the rope length variation parameter and the oscillation state parameter: The working condition of the hoisting task is identified based on the rope length change rate; If the working condition is a variable length release and take-up mode, the rope length change rate obtained in real time is used to participate in the calculation of the Coriolis feedforward compensation component. If the operating condition is constant altitude cruise mode, then the rope length change rate is set to 0 before participating in the calculation of the Coriolis feedforward compensation component.

[0007] In an alternative embodiment of the first aspect, the method further includes: If the rate of change of rope length is greater than the preset dead zone threshold, then the working condition is determined to be the variable length release / retraction mode. If the rate of change of the rope length is less than or equal to the preset dead zone threshold, then the working condition is determined to be constant altitude cruise mode.

[0008] In an alternative embodiment of the first aspect, mapping the updated position gain of the load position deviation to a position feedback guidance component in the velocity domain includes: Obtain the current position and desired position of the load, and calculate the load position deviation based on the current position and the desired position; The load position deviation is smoothed and filtered to obtain the static position offset; The difference between the load position deviation and the static position offset is mapped to a position feedback guidance component in the velocity domain based on the updated position gain.

[0009] In an optional embodiment of the first aspect, the rope length variation parameter includes the rope length variation rate, the oscillation state parameter includes the oscillation angular velocity, and the generation of Coriolis feedforward compensation components in the velocity domain based on the rope length variation parameter and the oscillation state parameter includes: Based on the rope length change rate, the swing angular velocity, and the Coriolis velocity mapping coefficient, a Coriolis feedforward compensation component in the velocity domain is generated.

[0010] In one alternative embodiment of the first aspect, acquiring real-time rope length data of the airborne winch to calculate rope length variation parameters includes: The absolute radian value of the airborne winch drum encoder is read, and the actual physical rope length is calculated by combining the drum's geometric parameters. The actual physical rope length is input to the tracking differentiator and smoothed and differentiated to extract the rope length change rate without phase hysteresis. The rope length acceleration is obtained by differentiating the rate of change of the rope length.

[0011] In an alternative embodiment of the first aspect, the method further includes: Extract the timestamps of the payload data frames to calculate the communication delay; If the communication delay meets the communication health requirements, the anti-sway control logic is activated: based on the real-time rope length update of position gain and velocity gain, position feedback guidance component, sway damping component and Coriolis feedforward compensation component are calculated and generated, and fused to obtain the target velocity offset vector.

[0012] In an alternative embodiment of the first aspect, the rope length variation parameter includes rope length acceleration, and the method further includes: Extract the timestamps of the payload data frames to calculate the communication delay; The detection of whether there is a load communication timeout and packet loss is based on the communication delay, and the detection of whether the load swing amplitude exceeds the preset safety limit is based on the swing state parameters; If a load communication timeout or packet loss is detected, the load swing amplitude exceeds the preset safety limit, or the rope length acceleration exceeds the preset critical acceleration threshold, all anti-swing control compensation values ​​are cleared, and an emergency brake command is issued to the airborne winch to drive the main unit out of the hoisting and escort flight state and switch to the original pure attitude wind-resistant hovering safety bottom-line working mode.

[0013] Secondly, this application provides a drone control system, comprising: The flight control main control unit is used to execute the UAV hoisting anti-sway control method based on variable rope length adaptive provided in the first aspect, so as to output the target velocity bias vector; A control bias fusion unit is used to receive the target velocity bias vector and superimpose the target velocity bias vector onto the flight control native target velocity node.

[0014] In an optional embodiment of the second aspect, the UAV control system further includes an encoder, which is mounted on an airborne servo winch and is used to collect position data corresponding to the current released rope length and output it to the flight control main control unit, so that the flight control main control unit can calculate the real-time rope length based on the position data.

[0015] In an optional embodiment of the second aspect, the UAV control system further includes a payload attitude sensor, which is disposed at the payload end and is used to collect payload attitude sensing data and transmit it to the flight control main control unit.

[0016] The UAV hoisting anti-sway control method and system based on variable rope length adaptive provided in this application have the following beneficial effects: By using real-time rope length as the independent variable, the velocity gain and position gain of the anti-sway control are updated nonlinearly; the updated position gain of the load position deviation is mapped to a position feedback guidance component in the velocity domain, and the updated velocity gain of the sway state parameters is mapped to a sway damping component in the velocity domain; and a Coriolis feedforward compensation component in the velocity domain is generated based on the rope length variation parameter and the sway state parameter; the position feedback guidance component, the sway damping component, and the Coriolis feedforward compensation component are fused to obtain the target velocity offset vector, which drives the fuselage to generate active following compensation motion. This application can solve the technical problem of anti-sway control failure and flight instability caused by nonlinear dynamic disturbances due to the dynamic time-varying rope length when existing multi-rotor UAVs perform winch hoisting operations. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] Figure 1 This is a structural block diagram of an unmanned aerial vehicle (UAV) control system provided in an embodiment of this application. Figure 2 This is a flowchart illustrating the UAV hoisting anti-sway control method based on variable rope length adaptive provided in the embodiments of this application. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0020] In related technologies, to suppress the swaying of the load during flight and wind disturbance, the "kinematic bias feedforward compensation" method is commonly used in the flight control field. Its basic principle is: by obtaining the sway angle and sway angular velocity of the load through sensors or state observers, the flight control system calculates a horizontal compensation amount (velocity bias or acceleration bias) that is opposite to the sway phase according to the control law, and feeds it forward into the UAV's desired position control loop, driving the multi-rotor main unit to actively generate following compensation motion, absorbing and dissipating the sway energy from the physical level.

[0021] However, the above method does not consider the dynamic changes in rope length. In actual hoisting operations with winches, the working rope length is usually not a constant value. The coupling between the rope release process and the load swing generates a significant Coriolis force, causing the system's dynamic characteristics to change in real time with the rope length. If the traditional fixed parameter control method is still used, it is impossible to predict and compensate for the drift of the pendulum's natural frequency and the Coriolis damping effect caused by the variable rope length. Ultimately, this leads to a sluggish response of the anti-swing system during the critical rope release stage of logistics delivery, and may even have a negative effect of reverse excitation, severely limiting the stability of the hoisting operation.

[0022] Furthermore, traditional anti-sway algorithms often exist only as an independent low-level control loop, lacking awareness and scheduling of the global operational status. In practical engineering applications, if the system encounters packet loss or high latency in the payload data link, or sudden extremely strong gusts causing the sway amplitude to exceed the limit, and lacks real-time multimodal task status identification and anomaly degradation mechanisms, it will continue to output speed bias based on erroneous data, seriously endangering flight safety.

[0023] To address at least one of the above technical problems, embodiments of this application provide a UAV hoisting anti-sway control method and a UAV control system based on variable rope length adaptive control.

[0024] like Figure 1 As shown, the UAV control system can be deployed on a multi-rotor UAV platform. The UAV control system includes at least a flight control main control unit 1 and a control bias fusion unit 2. The flight control main control unit 1 internally has two decoupled logic control modules: a state scheduling module 11 and an adaptive calculation module 12.

[0025] The state scheduling module 11 is used to run the task state machine and monitor the operating status of the airborne winch (including the working status of the hoisting task) and the health of the data communication link in real time. The state scheduling module 11 has built-in hysteresis comparison logic, which can avoid high-frequency jumps in the system at the critical point of mode switching and ensure the smooth and reliable mode transition.

[0026] The adaptive calculation module 12 is used to receive the gain scheduling parameters issued by the state scheduling module 11 and to complete the calculation of the position feedback guidance component, the swing damping component and the Coriolis feedforward compensation component in real time.

[0027] The state scheduling module 11 and the adaptive solution module 12 work together to implement the UAV hoisting anti-sway control method based on variable rope length adaptation provided in this embodiment, so as to output the target velocity bias vector to the control bias fusion unit 2.

[0028] The control bias fusion unit 2 adopts a classic cascaded feedforward injection architecture (i.e., position-velocity control loop). The target velocity bias vector is superimposed on the original target velocity node of the flight controller to ensure that even if the algorithm outputs a saturation extreme value, the fuselage attitude is limited by the system's physical safety threshold and does not disrupt the original navigation logic.

[0029] Optionally, the UAV control system also includes an encoder 3, which is mounted on the airborne servo winch. The encoder 3 collects position data corresponding to the current released rope length and outputs it to the flight control main unit 1, enabling the flight control main unit 1 to calculate the real-time rope length based on the position data corresponding to the current released rope length. Compared to the traditional method of indirectly estimating rope length based on GPS altitude difference, this application directly collects encoder data from the airborne winch to calculate the actual physical rope length, effectively improving the accuracy of rope length calculation.

[0030] Optionally, the UAV control system also includes a payload attitude sensor. The payload attitude sensor is located at the payload end and is used to collect payload attitude sensing data and transmit it to the flight control main unit.

[0031] This embodiment provides a drone hoisting anti-sway control method based on variable rope length adaptation. This scheme innovatively introduces a Coriolis force compensation term based on the rope length change rate into the control law. This feedforward compensation mechanism can counteract the additional sway caused by energy injection during winch load deployment and retrieval from a dynamic perspective, achieving synchronous and coordinated execution of rope deployment and retrieval operations, wind disturbance suppression, and load sway elimination, effectively improving hoisting operation efficiency.

[0032] Furthermore, this solution constructs a multi-modal mission state machine scheduling architecture that includes constant altitude cruise mode, variable length deployment and retraction mode, and abnormal degradation mode, overcoming the limitations of single control loop in adaptability to different operating conditions. When extreme conditions occur, such as data communication link timeout or load swing exceeding the safety threshold, the system can quickly isolate the anti-sway control loop and simultaneously execute the winch emergency brake and the aircraft's native pure attitude hovering control, forming an industrial-grade safety fallback mechanism to ensure flight operation safety under extreme conditions.

[0033] The following is a detailed description of the UAV hoisting and anti-sway control method based on variable rope length adaptive provided in the embodiments of this application, with reference to the accompanying drawings.

[0034] Please see Figure 2 The UAV hoisting anti-sway control method based on variable rope length adaptive provided in one embodiment of this application includes the following steps: S10, acquire real-time rope length data of the airborne winch to calculate rope length change parameters, and acquire load swing state parameters. S20, with the real-time rope length as the independent variable, nonlinearly update the velocity gain and position gain of the anti-swing control, and obtain the updated velocity gain and updated position gain accordingly; S30, the updated position gain of the load position deviation is mapped to the position feedback guidance component in the velocity domain, the updated velocity gain of the swing state parameters is mapped to the swing damping component in the velocity domain, and the Coriolis feedforward compensation component in the velocity domain is generated based on the rope length change parameter and the swing state parameter. S40 fuses the position feedback guidance component, the oscillation damping component, and the Coriolis feedforward compensation component to obtain the target velocity offset vector, and drives the fuselage to generate active following compensation motion based on the target velocity offset vector.

[0035] In this embodiment of the application, the rope length change parameters in step S10 include, but are not limited to, the rope length change rate and the rope length acceleration. The rope length change rate is the first derivative of the real-time rope length with respect to time, representing the amount of rope length change per unit time; the rope length acceleration is the first derivative of the rope length change rate with respect to time (i.e., the second derivative of the rope length with respect to time), representing the speed of the rope length change.

[0036] In existing hoisting anti-sway solutions, real-time rope length is usually obtained through estimation, such as indirectly estimating the rope length using the altitude difference from the aircraft's altimeter. This method is susceptible to interference from various factors, including aircraft attitude fluctuations, barometric altimeter drift, and changes in the working surface elevation. The rope length estimation accuracy is low, and the dynamic response is lagging, making it impossible to provide reliable state input for variable rope length anti-sway control.

[0037] To overcome the problem of insufficient rope length sensing accuracy caused by the height difference estimation method, this embodiment reads the output signal of the drum encoder through the airborne bus, combines the drum geometric parameters to calculate the real physical rope length, and calculates the rope length change rate and rope length acceleration based on the real physical rope length. This ensures the accuracy and real-time performance of rope length state sensing from the data source, providing more accurate input data for subsequent adaptive gain scheduling and Coriolis feedforward compensation.

[0038] Specifically, in this embodiment, obtaining real-time rope length data of the airborne winch to calculate rope length variation parameters includes the following steps: The absolute radian value of the airborne winch drum encoder is read, and the actual physical rope length is calculated by combining the drum's geometric parameters. The actual physical rope length is input into the tracking differentiator for smoothing filtering and differentiation, and the rope length change rate without phase hysteresis is extracted. The acceleration of the rope length is obtained by differentiating the rate of change of rope length.

[0039] In practical implementation, the zero-point calibration is first performed: when the system is powered on or the hook is fully retracted to the initial reset position, the absolute radian value corresponding to the initial reading of the encoder is recorded as the zero-point reference φ0. At this time, the hoisting system has an initial inherent bias rope length L0 (i.e., the physical distance from the winch outlet to the load center of mass).

[0040] Then, in actual drone hoisting operations, the absolute position of the drum encoder is periodically read via the onboard bus, and the zero-point reference φ0 is subtracted from it to obtain the absolute arc value of the current cumulative rope length. .

[0041] Finally, considering the effective winding radius R of the winch drum, the actual physical rope length L(t) at the current moment is calculated using the following formula: ×R.

[0042] This scheme uses geometric mapping between the position data of the drum encoder and the geometric parameters of the drum to calculate the true physical rope length more accurately. At the same time, to address the defect that direct differential of the encoder's discrete sampling signals can easily lead to high-frequency noise amplification, a tracking differentiator is introduced to smooth and filter the rope length sequence. This suppresses sampling noise interference while preserving the dynamic characteristics of the signal, further improving the accuracy and reliability of calculating the true physical rope length and its derived variation parameters.

[0043] Optionally, if the system is equipped with a load attitude sensor, the swing state data in S10 above may include the load attitude sensor data collected by the load attitude sensor. In this case, the flight control main control unit uses the extended Kalman filter algorithm based on the attitude sensor data uploaded by the load end (including attitude angle, angular velocity and position measurement value), and combines the motion data collected by the fuselage IMU to perform time alignment and forward prediction, compensate for the phase deviation caused by communication delay and packet loss, and calculate the current absolute swing angle and swing angular velocity of the load (i.e., the rate of change of the absolute swing angle with time).

[0044] Alternatively, if the load end is not equipped with a load attitude sensor, the swing state data in S10 is indirectly obtained from the motion parameters of the fuselage. The flight control main control unit extracts the measured acceleration of the fuselage, calculates the difference between it and the theoretical acceleration in the no-load state of the multi-rotor, and obtains the instantaneous acceleration residual of the fuselage; combined with the total thrust of the motor, the horizontal tension component of the rope on the fuselage is solved, and combined with the real physical rope length obtained in real time, the current absolute swing angle and swing angular velocity of the load are obtained by reverse observation through the thrust-attitude coupling observer.

[0045] It should be noted that the methods for calculating rope length, absolute swing angle, and swing angular velocity described in the foregoing embodiments are merely illustrative examples and do not constitute a limitation on the scope of protection of this application. In specific implementations, existing methods for estimating rope length based on height difference and methods for obtaining swing angle based on load sensors can also be used.

[0046] Optionally, after obtaining the rate of change of rope length and the acceleration of rope length, the flight control main control unit can perform the working condition identification of the hoisting task: If the rate of change of rope length is greater than the preset dead zone threshold, the current operating condition is determined to be the variable length release / retraction mode; if the rate of change of rope length is less than or equal to the preset dead zone threshold, the current operating condition is determined to be the constant altitude cruise mode. In the constant altitude cruise mode, the rate of change of rope length is assigned a value of 0.

[0047] It should be noted that this application distinguishes between the variable-length extension / retraction mode and the constant-altitude cruise mode because, in actual flight operations, multi-rotor fuselages experience severe inherent mechanical vibrations. When the airborne winch is physically locked, the drum encoder, coupled with the fuselage vibration, will still output minute glitch-like jump signals, thereby calculating a false time-varying rope length change rate. If the independent constant-altitude cruise mode is not isolated through the dead-zone threshold, this type of high-frequency noise will be directly input into the subsequent control law calculation stage, causing the flight control system to continuously output meaningless minute speed offsets, resulting in abnormal fuselage vibration.

[0048] After classifying the hoisting conditions using dead-zone thresholds, differentiated calculations are performed for different modes in subsequent gain calculations and control quantity generation: in the variable-length extension / retraction mode, the real-time calculated rope length change rate is used in the calculation of the Coriolis feedforward compensation component; in the constant-altitude cruise mode, the rope length change rate is first set to 0 before being used in the calculation of the Coriolis feedforward compensation component. Thus, the Coriolis feedforward compensation component based on the rope length change rate in the constant-altitude cruise mode will automatically return to zero, effectively suppressing vibration and noise interference in the steady-state phase. This is a key adaptive hysteresis mechanism to ensure the system's steady-state noise immunity.

[0049] When the system is in variable-length extension / retraction mode or constant-altitude cruise mode, the flight control unit executes the anti-sway control logic. Specifically, the flight control unit uses the real-time physical cable length as the independent variable to nonlinearly update the velocity gain and position gain of the anti-sway control, obtaining the updated velocity gain and position gain. Among these, the velocity gain... The update formula is as follows: ;

[0050] in, The reference velocity gain is a dimensionless proportional mapping coefficient between the load swing velocity and the flight control target velocity correction obtained through parameter tuning and optimization during the ground calibration phase under the reference rope length condition. L(t) represents the actual physical rope length at time t, in meters. The preset reference rope length is in meters (m).

[0051] Position gain The update formula is as follows: ;

[0052] in, The reference position gain is the control parameter obtained during the ground calibration phase under the reference reference rope length condition. It represents the magnitude of the velocity correction corresponding to a unit position deviation (in the position-velocity cascade control architecture of the flight controller, the control logic matches the corresponding output velocity correction based on the input position deviation magnitude). The dimension of this parameter is 1 / s.

[0053] Optionally, mapping the updated position gain of the load position deviation to a position feedback guidance component in the velocity domain includes the following steps: Obtain the current position and desired position of the load, and calculate the load position deviation based on the current position and desired position; The load position deviation is smoothed and filtered to obtain the static position offset; The difference between the load position deviation and the static position offset is mapped to a position feedback guide component in the velocity domain based on the updated position gain.

[0054] More specifically, position feedback guidance component The calculation formula is as follows: ; in, For the desired position, Current position This is a static position offset. , , The unit for all of them is m.

[0055] Optionally, the updated velocity gain of the oscillation state parameters can be mapped to the oscillation damping component in the velocity domain using the following calculation formula. : ; in, This represents the horizontal velocity component of the load due to its oscillation, expressed in m / s.

[0056] Optionally, the Coriolis feedforward compensation component is determined by the following steps: The Coriolis feedforward compensation component in the velocity domain is generated based on the rope length change rate, load swing angular velocity, and Coriolis velocity mapping coefficient. The calculation formula is as follows: ; Where ∆L(t) is the rate of change of the rope length over time t, in m / s. The angular velocity of the load swing at time t is expressed in rad / s. For the Coriolis feedforward compensation component, This is the Coriolis velocity mapping coefficient (i.e., the coefficient in the Coriolis force), with units of rad / s.

[0057] Finally, the position feedback guidance component, the oscillation damping component, and the Coriolis feedforward compensation component are added together to obtain the target velocity offset vector.

[0058] Then, the target velocity offset vector is input to the control offset fusion unit. Based on the position-velocity control loop, the control offset fusion unit superimposes the target velocity offset vector onto the target velocity node output by the original position loop to complete the correction, thereby obtaining the final expected velocity command for fusion anti-sway compensation. The final expected velocity command is then sent to the underlying velocity control loop to drive the fuselage to generate the corresponding compensation motion.

[0059] The UAV hoisting anti-sway control method based on variable rope length adaptive provided in this application embodiment overcomes the application limitations of fixed gain control. It directly obtains the high-precision real physical rope length through an onboard winch. Based on the dynamic law of the pendulum's natural frequency changing with rope length, it adaptively adjusts the controller's gain coefficient in real time, ensuring that the UAV maintains the optimal system damping ratio under any hovering altitude and rope length conditions, effectively suppressing fuselage control overshoot and response hysteresis. Furthermore, this embodiment innovatively introduces a Coriolis feedforward compensation term based on the rope length change rate into the control law, offsetting the additional sway caused by energy injection during winch load deployment and retraction from the root of dynamics, achieving efficient coordination of rope deployment, wind resistance stability, and sway suppression.

[0060] This application also provides another embodiment of the UAV hoisting anti-sway control method based on variable rope length adaptive. The difference between this embodiment and the previous embodiment is that this embodiment also considers the fault-tolerant processing mechanism for multi-source sensor failure and communication anomaly scenarios.

[0061] Specifically, in this embodiment, the UAV hoisting anti-sway control method based on variable rope length adaptive may further include: Extract the timestamp of the payload data frame transmission and the timestamp of the corresponding payload data frame received by the flight controller to calculate the communication delay.

[0062] Specifically, the payload and the flight control master are two independent embedded systems, each equipped with its own local clock. The communication delay is calculated by subtracting the transmission timestamp carried in the payload's pose data frame from the reception timestamp of the data received by the flight control master.

[0063] In this embodiment, the triggering condition for the sway control logic also includes: the communication delay meeting a preset communication health requirement. For example, when the communication delay is less than a preset delay tolerance threshold, it is determined that the communication health requirement is met; otherwise, it is determined that there is a load communication timeout and packet loss.

[0064] Optionally, the UAV hoisting anti-sway control method based on variable rope length adaptive provided in the embodiments of this application may further include: The system detects whether there is a timeout or packet loss in the load communication based on communication delay, whether the load swing amplitude exceeds the preset safety limit based on swing state parameters, and whether the rope length acceleration is greater than the preset critical acceleration threshold.

[0065] If a load communication timeout or packet loss is detected, the load swing exceeds the preset safety limit, or the rope length acceleration exceeds the preset critical acceleration threshold, then a hardware-level data mutation or strong electromagnetic interference is determined to have occurred. The state machine forcibly blocks other logic and switches to the abnormal degradation mode.

[0066] When the system is in an abnormal degraded mode, the system forcibly clears all anti-sway control compensation quantities (i.e., position feedback guidance components, sway damping components and Coriolis feedforward compensation components), and issues an emergency brake command to the airborne winch, driving the main unit to exit the hoisting and escort flight state and switch to the original pure attitude wind-resistant hovering safety protection working mode.

[0067] Specifically, if the absolute swing angle of the load is greater than or equal to the preset safety threshold, then the load swing amplitude is determined to exceed the preset safety limit.

[0068] This embodiment abandons the unidirectional control architecture and constructs a multimodal task state machine that includes fixed-length cruise, variable-length extension and retraction, and abnormal degradation. When encountering extreme conditions such as data link timeout or load swing exceeding the limit, the system can quickly isolate the swing control loop and simultaneously execute winch braking and pure attitude hovering of the fuselage, providing industrial-grade safety backup capability.

[0069] This application also provides another embodiment of the UAV hoisting anti-sway control method based on variable rope length adaptive control. The difference between this embodiment and the above embodiment is that, when the system is detected to be in constant altitude cruise mode, the flight control main control unit can also execute the following anti-sway control logic: First, based on the actual physical rope length L(t), the theoretical natural resonant frequency under the current rope length condition is calculated using the simple pendulum formula. The calculation formula is: , This is the acceleration due to gravity.

[0070] Then, a time-sliding window (e.g., a continuous data stream of the past 3 to 5 seconds) is opened for the current absolute swing angle or angular velocity of the load. A zero-crossing rate (ZCR) algorithm or a lightweight fast Fourier transform (FFT) is run within the MCU to extract the current dominant frequency of the load under external force excitation in real time. .

[0071] When the system detects at high frequency that the true disturbance frequency is close to the natural frequency of the pendulum (i.e., it satisfies...) , When the load's current absolute swing angle exceeds the set safety boundary (where the value is the preset minimum), and physical resonance is detected, a step or pulse micro-motion command is immediately sent to the servo winch. This is achieved by retrieving a pre-calibrated and... Matching rope length mapping table, changing rope length L to forcibly destroy The boundary physical conditions enable the active separation of the three-dimensional spatial cable from resonance.

[0072] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Modules described in the embodiments of this application can be implemented in software or hardware. The names of modules do not, in some cases, constitute a limitation on the unit itself.

[0073] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for anti-sway control of UAV hoisting based on variable rope length adaptive control, characterized in that, include: To obtain real-time rope length data of the airborne winch in order to calculate rope length variation parameters, and to obtain load swing state parameters; Using the real-time rope length as the independent variable, the velocity gain and position gain of the anti-swing control are updated nonlinearly, and the updated velocity gain and position gain are obtained accordingly. The updated position gain of the load position deviation is mapped to the position feedback guidance component in the velocity domain, the updated velocity gain of the swing state parameter is mapped to the swing damping component in the velocity domain, and the Coriolis feedforward compensation component in the velocity domain is generated based on the rope length change parameter and the swing state parameter. The position feedback guidance component, the swing damping component, and the Coriolis feedforward compensation component are fused to obtain the target velocity offset vector, and the fuselage is driven to generate active following compensation motion based on the target velocity offset vector.

2. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 1, characterized in that, The rope length variation parameter includes the rope length variation rate, and before generating the Coriolis feedforward compensation component in the velocity domain based on the rope length variation parameter and the swing state parameter, the following is also included: The working condition of the hoisting task is identified based on the rope length change rate; If the working condition is a variable length release and take-up mode, the rope length change rate obtained in real time is used to participate in the calculation of the Coriolis feedforward compensation component. If the operating condition is constant altitude cruise mode, then the rope length change rate is set to 0 before participating in the calculation of the Coriolis feedforward compensation component.

3. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 2, characterized in that, The method further includes: If the rate of change of rope length is greater than the preset dead zone threshold, then the working condition is determined to be the variable length release / retraction mode. If the rate of change of the rope length is less than or equal to the preset dead zone threshold, then the working condition is determined to be constant altitude cruise mode.

4. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 1, characterized in that, The step of mapping the updated position gain of the load position deviation to a position feedback guidance component in the velocity domain includes: Obtain the current position and desired position of the load, and calculate the load position deviation based on the current position and the desired position; The load position deviation is smoothed and filtered to obtain the static position offset; The difference between the load position deviation and the static position offset is mapped to a position feedback guidance component in the velocity domain based on the updated position gain.

5. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 1, characterized in that, The rope length variation parameter includes the rope length variation rate, the swing state parameter includes the swing angular velocity, and the generation of Coriolis feedforward compensation components in the velocity domain based on the rope length variation parameter and the swing state parameter includes: Based on the rope length change rate, the swing angular velocity, and the Coriolis velocity mapping coefficient, a Coriolis feedforward compensation component in the velocity domain is generated.

6. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 1, characterized in that, The process of acquiring real-time rope length data of the airborne winch to calculate rope length variation parameters includes: The absolute radian value of the airborne winch drum encoder is read, and the actual physical rope length is calculated by combining the drum's geometric parameters. The actual physical rope length is then input into a tracking differentiator for smoothing filtering and differentiation processing to extract the rope length change rate without phase hysteresis. The rope length acceleration is obtained by differentiating the rate of change of the rope length.

7. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 1, characterized in that, The method further includes: Extract the timestamps of the payload data frames to calculate the communication delay; If the communication delay meets the communication health requirements, the anti-sway control logic is activated: based on the real-time rope length update of position gain and velocity gain, position feedback guidance component, sway damping component and Coriolis feedforward compensation component are calculated and generated, and fused to obtain the target velocity offset vector.

8. The UAV hoisting anti-sway control method based on variable rope length adaptive as described in claim 1, characterized in that, The rope length variation parameter includes rope length acceleration, and the method further includes: Extract the timestamps of the payload data frames to calculate the communication delay; The detection of whether there is a load communication timeout and packet loss is based on the communication delay, and the detection of whether the load swing amplitude exceeds the preset safety limit is based on the swing state parameters; If a load communication timeout or packet loss is detected, the load swing amplitude exceeds the preset safety limit, or the rope length acceleration exceeds the preset critical acceleration threshold, all anti-swing control compensation values ​​are cleared, and an emergency brake command is issued to the airborne winch to drive the main unit out of the hoisting and escort flight state and switch to the original pure attitude wind-resistant hovering safety bottom-line working mode.

9. A drone control system, characterized in that, include: The flight control main control unit is used to execute the UAV hoisting anti-sway control method based on variable rope length adaptive as described in any one of claims 1 to 8, so as to output the target velocity bias vector; A control bias fusion unit is used to receive the target velocity bias vector and superimpose the target velocity bias vector onto the flight control native target velocity node.

10. The unmanned aerial vehicle control system as described in claim 9, characterized in that, The UAV control system also includes an encoder, which is mounted on an airborne servo winch and is used to collect position data corresponding to the current released rope length and output it to the flight control main control unit so that the flight control main control unit can calculate the real-time rope length based on the position data. And / or, the UAV control system further includes a payload attitude sensor, which is disposed at the payload end and is used to collect payload attitude sensing data and transmit it to the flight control main control unit.

Images