A wind-resistant control method and system for a flyable robot

Abstract

The application relates to the technical field of unmanned plane control, and discloses a wind-resistant control method and system for a flying robot. According to the application, multi-dimensional sensing detection data are acquired in real time and combined with a state prediction equation, so that a deviation caused by external wind disturbance can be accurately separated; secondly, an adaptive law is used to estimate the prediction error, and a wind disturbance torque matched with the current wind disturbance is generated; thirdly, a filtering operation and a feedforward compensation calculation are performed on the wind disturbance torque, so that the compensation amount is converted into accurate target feedforward compensation amount, and the interference of measurement noise and estimation error is effectively inhibited; finally, the target feedforward compensation amount is fused with a baseline control amount and converted into a motor speed instruction, feedforward-feedback compound control of wind disturbance is realized, the effectiveness of original trajectory and attitude correction functions is maintained, the active inhibition capability of wind disturbance is increased, and the flight stability and wind resistance of the robot are improved.

Description

Technical Field

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

[0002] With the rapid development of the photovoltaic power generation industry, the demand for cleaning and maintenance of photovoltaic power plants is increasing. Flying robots, as a new type of equipment, can hover above photovoltaic panels to perform non-contact cleaning operations, and have advantages such as high work efficiency and no limitation by ground terrain.

[0003] However, these robots typically need to hover at a certain height above the ground, making them highly susceptible to interference from environmental wind fields. Existing flight control methods for these robots mainly employ traditional control strategies such as PID control or linear quadratic regulators, using feedback adjustment to correct attitude deviations. However, in practical applications, due to the randomness, abrupt changes, and nonlinearity of wind disturbances, relying solely on feedback control suffers from response lag. Furthermore, when encountering gusts or persistent crosswinds, the robot often exhibits significant positional drift or attitude jitter, and in severe cases, uncontrollable collisions with solar panels or crashes may occur.

[0004] While some research has attempted to introduce wind disturbance observers for compensation control, existing observation methods often treat wind disturbances as fixed disturbances or slowly changing quantities, making it difficult to accurately capture the real-time characteristics of wind disturbances and resulting in limited compensation effects. Therefore, achieving precise and rapid wind resistance control in complex and variable wind field environments has become a critical issue that urgently needs to be addressed in the development of flyable robot technology. Summary of the Invention

[0005] This invention provides a wind-resistant control method and system for a flying robot, which can achieve precise wind-resistant control of the flying robot, enabling it to sense changes in wind disturbance in real time and actively generate matching compensation torque, thereby ensuring its flight attitude and cleaning operation trajectory.

[0006] The first aspect of this invention discloses a wind-resistant control method for a flyable robot, the method comprising: The system acquires real-time sensor detection data for the flyable robot and calculates the prediction error of the flyable robot based on the sensor detection data and a pre-constructed state prediction equation. The state prediction equation is used to predict the flight data of the flyable robot in an ideal state. The sensor detection data includes at least sub-detection data corresponding to multiple preset state vectors. The prediction error is estimated according to a preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error; and the wind disturbance moment is filtered and fedforward compensation is calculated to obtain the target feedforward compensation amount corresponding to the wind disturbance moment. The baseline control quantity of the flyable robot and the target feedforward compensation quantity are converted into motor speed commands, and the flyable robot is controlled to execute the motor speed commands to adjust the flight parameters of the flyable robot; the baseline control quantity is used to correct the trajectory and / or attitude of the flyable robot.

[0007] As an optional implementation, in the first aspect of the present invention, the method further includes: A dynamic equation is constructed for a flyable robot, which is described by a six-degree-of-freedom system including three translational degrees of freedom and three rotational degrees of freedom. The dynamic equations are converted into state prediction equations and initialized. The state prediction equations include multiple state vectors and multiple control input vectors. The multiple state vectors include at least a first sub-vector corresponding to the body's center of mass position, velocity, attitude angle, and angular velocity. The multiple control input vectors include at least a second sub-vector corresponding to the total thrust, roll moment, pitch moment, and yaw moment.

[0008] As an optional implementation, in the first aspect of the present invention, the step of calculating the prediction error of the flyable robot based on the sensor detection data and the pre-constructed state prediction equation includes: The difference between the state prediction equation and the dynamic equation is calculated to obtain the prediction error dynamic equation; According to a preset fusion algorithm, a data fusion operation is performed on all the sub-detection data included in the sensing detection data to obtain a data fusion result corresponding to all the sub-detection data; the preset fusion algorithm includes a fusion algorithm based on a Kalman filter; the data fusion result is a multi-dimensional fusion vector corresponding to all the state vectors; Substituting the data fusion result into the prediction error dynamic equation yields the prediction error of the flyable robot.

[0009] As an optional implementation, in the first aspect of the present invention, the step of performing filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain a target feedforward compensation amount corresponding to the wind disturbance moment includes: A low-pass filtering operation is performed on the wind disturbance torque according to a preset low-pass filter to obtain a low-pass filtering result corresponding to the wind disturbance torque; Based on the preset feedforward compensation calculation formula and combined with the preset rate saturation constraint, the target feedforward compensation amount for the low-pass filtering result is calculated; wherein, the rate saturation constraint is used to limit the rate of change of the target feedforward compensation amount.

[0010] As an optional implementation, in the first aspect of the present invention, the feedforward compensation calculation formula is:

[0011] in, This is the feedforward compensation amount for the target; This is the feedforward gain matrix; the negative sign indicates that the compensation direction of the target feedforward compensation is opposite to the wind resistance direction. This is the wind disturbance moment.

[0012] As an optional implementation, in the first aspect of the present invention, the kinetic equation is:

[0013] in, ,and This describes the system status of the flying robot. Baseline control input; For adaptive control input; To match uncertainty, This represents non-matching uncertainty; These are the uncertainty distribution matrices for matched and unmatched products, respectively; The state prediction equation is:

[0014] in, ,and The state corresponding to the preset predictor. To address this uncertainty The estimate; , This represents the prediction error; , and is a Hurwitz matrix; For the augmented distribution matrix, and .

[0015] As an optional implementation, in the first aspect of the present invention, the prediction error dynamic equation is:

[0016] in, And it is a generalized uncertainty term, in which, To match uncertainty.

[0017] A second aspect of the present invention discloses a wind-resistant control system for a flyable robot, the system comprising: The data acquisition module is used to acquire sensor detection data for the flying robot in real time; An error calculation module is used to calculate the prediction error of the flyable robot based on the sensor detection data and a pre-constructed state prediction equation; the state prediction equation is used to predict the flight data of the flyable robot in an ideal state; the sensor detection data includes at least sub-detection data corresponding to multiple preset state vectors; The data estimation module is used to perform data estimation on the prediction error according to a preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error; The feedforward compensation calculation module is used to perform filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment. The instruction conversion and execution module is used to convert the baseline control quantity and the target feedforward compensation quantity of the flyable robot into motor speed commands, and control the flyable robot to execute the motor speed commands to adjust the flight parameters of the flyable robot; the baseline control quantity is used to correct the trajectory and / or attitude of the flyable robot.

[0018] As an optional implementation, in a second aspect of the invention, the system further includes: A building module is used to construct the dynamic equations for a flyable robot, which are described using a six-degree-of-freedom approach including three translational degrees of freedom and three rotational degrees of freedom. The equation transformation module is used to convert the dynamic equations into state prediction equations and initialize the state prediction equations. The state prediction equations include multiple state vectors and multiple control input vectors. The multiple state vectors include at least a first sub-vector corresponding to the body's center of mass position, velocity, attitude angle, and angular velocity. The multiple control input vectors include at least a second sub-vector corresponding to the total thrust, roll moment, pitch moment, and yaw moment.

[0019] As an optional implementation, in a second aspect of the present invention, the error calculation module calculates the prediction error of the flyable robot based on the sensor detection data and a pre-constructed state prediction equation in the following specific ways: The difference between the state prediction equation and the dynamic equation is calculated to obtain the prediction error dynamic equation; According to a preset fusion algorithm, a data fusion operation is performed on all the sub-detection data included in the sensing detection data to obtain a data fusion result corresponding to all the sub-detection data; the preset fusion algorithm includes a fusion algorithm based on a Kalman filter; the data fusion result is a multi-dimensional fusion vector corresponding to all the state vectors; Substituting the data fusion result into the prediction error dynamic equation yields the prediction error of the flyable robot.

[0020] As an optional implementation, in the second aspect of the present invention, the feedforward compensation calculation module performs filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment, specifically including: A low-pass filtering operation is performed on the wind disturbance torque according to a preset low-pass filter to obtain a low-pass filtering result corresponding to the wind disturbance torque; Based on the preset feedforward compensation calculation formula and combined with the preset rate saturation constraint, the target feedforward compensation amount for the low-pass filtering result is calculated; wherein, the rate saturation constraint is used to limit the rate of change of the target feedforward compensation amount.

[0021] As an optional implementation, in a second aspect of the present invention, the feedforward compensation calculation formula is:

[0022] in, This is the feedforward compensation amount for the target; This is the feedforward gain matrix; the negative sign indicates that the compensation direction of the target feedforward compensation is opposite to the wind resistance direction. This is the wind disturbance moment.

[0023] As an optional implementation, in a second aspect of the invention, the kinetic equation is:

[0024] in, ,and This describes the system status of the flying robot. Baseline control input; For adaptive control input; To match uncertainty, This represents non-matching uncertainty; These are the uncertainty distribution matrices for matched and unmatched products, respectively; The state prediction equation is:

[0025] in, ,and The state corresponding to the preset predictor. To address this uncertainty The estimate; , This represents the prediction error; , and is a Hurwitz matrix; For the augmented distribution matrix, and .

[0026] As an optional implementation, in a second aspect of the invention, the prediction error dynamic equation is:

[0027] in, And it is a generalized uncertainty term, in which, To match uncertainty.

[0028] A third aspect of the present invention discloses a wind-resistant control device for a flyable robot, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the wind-resistant control method for the flyable robot according to any of the first aspects of the present invention.

[0029] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the wind-resistant control method for any of the flying robots described in the first aspect of the present invention.

[0030] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a wind-resistant control method for a flyable robot. The method includes: acquiring real-time sensor detection data of the flyable robot, and calculating the prediction error of the flyable robot based on the sensor detection data and a pre-constructed state prediction equation; the state prediction equation is used to predict the flight data of the flyable robot under ideal conditions; the sensor detection data includes at least sub-detection data corresponding to multiple preset state vectors; performing data estimation on the prediction error according to a preset adaptive law to obtain the wind disturbance torque corresponding to the prediction error; performing filtering and feedforward compensation calculation operations on the wind disturbance torque to obtain a target feedforward compensation amount corresponding to the wind disturbance torque; converting the baseline control quantity and the target feedforward compensation amount of the flyable robot into motor speed commands, and controlling the flyable robot to execute the motor speed commands to adjust the flight parameters of the flyable robot; the baseline control quantity is used to correct the trajectory and / or attitude of the flyable robot. As can be seen, by implementing this invention, firstly, by acquiring multi-dimensional sensor data in real time and combining it with the state prediction equation, the deviation caused by external wind disturbance can be accurately separated, which is beneficial to improving the accuracy of wind disturbance identification; secondly, by using an adaptive law to estimate the prediction error, a wind disturbance torque that matches the current wind disturbance can be generated in real time, improving the robot's adaptability to sudden changes in wind speed and direction; thirdly, by performing filtering operations and feedforward compensation calculations on the wind disturbance torque, the compensation amount is transformed into a precise target feedforward compensation amount, effectively suppressing the interference of measurement noise and estimation errors, and improving the smoothness and stability of the subsequently generated motor speed commands; finally, by fusing the target feedforward compensation amount with the baseline control amount and converting it into a motor speed command, a feedforward-feedback composite control for wind disturbance is realized, which not only maintains the effectiveness of the original trajectory and attitude correction functions, but also increases the active suppression capability of wind disturbance, which is beneficial to improving the robot's flight stability and wind resistance performance when operating at high altitudes. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a structural schematic diagram of a flying robot disclosed in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a wind-resistant control method for a flyable robot disclosed in an embodiment of the present invention. Figure 3 This is a flowchart illustrating another wind-resistant control method for a flying robot disclosed in an embodiment of the present invention; Figure 4 This is a schematic diagram of the wind-resistant control system for a flying robot disclosed in an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of another wind-resistant control system for a flying robot disclosed in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a wind-resistant control device for a flying robot disclosed in an embodiment of the present invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0035] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0036] This invention discloses a wind-resistant control method and system for a flyable robot. First, by acquiring multi-dimensional sensor data in real time and combining it with state prediction equations, the deviation caused by external wind disturbances can be accurately separated, improving the accuracy of wind disturbance identification. Second, by employing an adaptive law to estimate the prediction error, a wind disturbance torque matching the current wind disturbance can be generated in real time, improving the robot's adaptability to sudden changes in wind speed and direction. Third, by performing filtering and feedforward compensation calculations on the wind disturbance torque, the compensation amount is transformed into a precise target feedforward compensation amount, effectively suppressing the interference of measurement noise and estimation errors, and improving the smoothness and stability of the subsequently generated motor speed commands. Finally, the target feedforward compensation amount is fused with the baseline control amount and converted into a motor speed command, realizing feedforward-feedback composite control of wind disturbances. This maintains the effectiveness of the original trajectory and attitude correction functions while increasing the active suppression capability of wind disturbances, which is beneficial to improving the robot's flight stability and wind resistance performance during high-altitude operations. Detailed explanations follow.

[0037] To better understand the wind resistance control method and system for the flyable robot described in this invention, the application structure to which the wind resistance control method for the flyable robot is applicable is first described. Specifically, this application structure can be as follows: Figure 1 As shown, Figure 1 This is a structural schematic diagram of a flying robot disclosed in an embodiment of the present invention. Figure 1 As shown, the application structure may include: 1. Tethering Module 2 (one at the front and one at the rear, also serving as a landing support), 3. Robotic Walking and Flying Unit, 4. Flying Wing Arms, 5. Flying Propeller, 6. Cleaning Brush, 7. Secondary Anti-Fall Radar, 8. Flying Gimbal Camera, 9. Walking Tracks, and 10. Primary Anti-Fall Radar. Tether 1 provides the robot with a continuous power supply and data transmission connection, ensuring that the robot can work at high altitudes for extended periods without frequently returning to the ground to recharge. The roller brush module 2, located at the front and rear of the robot, not only has a cleaning function, but also serves as a support and cushioning mechanism when the flying robot lands.

[0038] Robot Walking and Flying Unit 3 is the core control unit of the robot, integrating a control system and power distribution for both walking and flying modes.

[0039] Flight wing arm 4 is a structural component that connects the fuselage and the flight propeller. It is used to support the flight system and can provide a certain amount of lift or stability during flight.

[0040] The fifth propeller provides the robot with flight power, enabling it to hover, rise and fall, and move forward, backward, left, and right.

[0041] The cleaning roller brush 6 is a component that performs specific cleaning tasks, sweeping away dust and debris from the floor or window surface by rotating the bristles.

[0042] The secondary fall protection radar 7 serves as a second line of defense, used to detect edges or suspended areas at higher positions to prevent robots from falling during walking or flying transitions.

[0043] The Flying Gimbal Camera 8 is a high-definition camera device installed on the flying section, used for aerial photography, environmental monitoring, or remote control to assist in visual perception.

[0044] The walking track 9 is a device that enables the robot to move on the ground or complex surfaces, providing strong grip and obstacle-crossing capabilities.

[0045] The primary anti-fall radar 10, located at a lower position, is the first detection device used to monitor the ground edge in real time to prevent the robot from falling while walking.

[0046] in, Figure 1 The flying robot shown can adopt the structure of a quadcopter drone, and can be applied to the cleaning of photovoltaic panels to achieve contactless cleaning of photovoltaic panels; optionally, the flying robot can also be applied to other cleaning scenarios, such as the cleaning of windows / exterior walls of high-rise buildings, etc., and the present invention does not limit it.

[0047] It should be noted that, Figure 1 The application structure shown is only to illustrate the equipment structure applicable to the wind resistance control method of the flying robot. The structures involved, such as tethering 1, roller brush module 2, robot walking and flying unit 3, flying wing arm 4, flying propeller 5, cleaning roller brush 6, secondary anti-fall radar 7, flying gimbal camera 8, walking track 9, and primary anti-fall radar 10, are only schematic representations. The specific structure / size / shape / location / installation method can be adapted to meet actual usage requirements. Figure 1 The application architecture shown is not limited in this respect.

[0048] The above describes the application structure to which the wind resistance control method for flying robots is applicable. The following is a detailed description of the wind resistance control method and system for flying robots.

[0049] Example 1 Please see Figure 2 , Figure 2 This is a flowchart illustrating a wind-resistant control method for a flyable robot disclosed in an embodiment of the present invention. Figure 2The described wind-resistant control method for flying robots can be applied to wind-resistant control devices / systems for flying robots, and the embodiments of this invention are not limited thereto. Figure 2 As shown, the wind resistance control method for this flyable robot may include the following operations: 101. Real-time acquisition of sensor detection data for flying robots.

[0050] 102. Based on the sensor detection data and the pre-constructed state prediction equation, the prediction error of the flying robot is calculated.

[0051] In this embodiment of the invention, the state prediction equation is used to predict the flight data of a flyable robot (hereinafter referred to as the robot) in an ideal state (the ideal state refers to the state without wind disturbance).

[0052] In this embodiment of the invention, the sensing detection data includes at least sub-detection data corresponding to multiple preset state vectors.

[0053] In this embodiment of the invention, by executing steps 101-102, the deviation caused by external wind disturbance can be accurately separated. This is different from the fuzzy processing method in traditional methods that simply classify wind disturbance as system noise, and can help improve the accuracy of wind disturbance identification.

[0054] In this embodiment of the invention, steps 101-102 and subsequent steps 103-105 can all be executed by the L1 adaptive controller built into the flying robot. The L1 adaptive controller mainly includes a state predictor and a low-pass filter. The L1 adaptive controller is equipped with an adaptive law and a feedforward compensation algorithm for performing feedforward compensation calculation.

[0055] 103. Perform data estimation on the prediction error according to the preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error.

[0056] In this embodiment of the invention, an adaptive law is used to estimate the prediction error, enabling the generation of wind disturbance torque that matches the current wind disturbance in real time. This dynamic compensation mechanism based on the adaptive law allows the robot to adapt to sudden changes in wind speed and direction, effectively solving the lag and mismatch problems of conventional fixed parameter compensation methods in complex wind field environments.

[0057] In this embodiment of the invention, the adaptive law employs a continuous-time adaptive law to update the parameter estimates in real time. For the estimation of time-varying parameters and disturbances, the general form of the adaptive law is:

[0058]

[0059]

[0060] in, For adaptive gain, This is a projection operator used to ensure that the estimated parameters remain within known conservative boundaries. , , These are the estimated values ​​for the unknown parameters, time-varying parameters, and disturbances, respectively. In this embodiment of the invention, considering the complexity of wind disturbance in the specific application of flyable robots, the "piecewise constant adaptive law" is more effective. For: T s Sampling time, ,and:

[0061] in, .

[0062] Accordingly, the uncertainty of the estimate is:

[0063] In this embodiment of the invention, the design concept of the piecewise constant adaptive law is to drive the prediction error to a minimum within each sampling period. When t = iT s When executing the adaptive law, terms caused by the initial conditions of the sampling period are eliminated. This resets the prediction error at the beginning of each sampling period. In this embodiment of the invention, it should be noted that the calculation of the wind disturbance moment integrates the prediction error, the parameter update of the adaptive law, and the processing of the low-pass filter. The complete calculation process of the wind disturbance moment is as follows: First, an estimate of the total uncertainty is obtained through the state predictor and the adaptive law: The estimate of this total uncertainty includes the impact of wind disturbance. According to the L1Quad architecture, the estimated uncertainty can be expressed as:

[0064] in, To match uncertainty estimation, This is an estimation of mismatched uncertainties. For calculating wind disturbance moments, it is necessary to extract the wind disturbance-related components from the total estimated uncertainty. It is assumed that wind disturbance can be modeled as a vector of generalized forces and moments. Then, in the body coordinate system, the formula for calculating the wind disturbance moment τw is:

[0065] Among them, Ω wLet J be the angular velocity disturbance caused by wind, and J be the body inertia matrix. In the L1 adaptive control architecture, the final control input is the superposition of baseline control and adaptive compensation:

[0066] Among them, baseline control u b (t) is calculated by the geometric controller or other nominal controller, u ad (t) is the output of the L1 adaptive controller, which includes a compensation term for wind disturbance torque.

[0067] In this embodiment of the invention, regarding the calculation of wind disturbance moment, considering that wind disturbance mainly affects the attitude dynamics of the UAV, the L1 adaptive controller primarily compensates for the uncertainty of the rotation path. Therefore, the formula for calculating the wind disturbance moment can be written as:

[0068] in, To represent the estimated wind disturbance torque in the frequency domain, C(s) is a low-pass filter, τ ad This is the final wind disturbance moment compensation term. In a continuous-time system, the time-domain expression for the wind disturbance moment is:

[0069] In this embodiment of the invention, the integral form shows that the wind disturbance moment is a weighted average of past estimates, with the weights decaying exponentially over time. At the same time, it also reflects the characteristics of low-pass filtering.

[0070] 104. Perform filtering and feedforward compensation calculations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment.

[0071] In this embodiment of the invention, by executing step 104, the compensation amount can be converted into an accurate target feedforward compensation amount, which effectively suppresses the interference of measurement noise and estimation error, and ensures the smoothness and stability of the compensation command (subsequent motor speed command). At the same time, the accurate calculation of the target feedforward compensation amount can also effectively avoid flight attitude oscillation caused by compensation signal jitter.

[0072] 105. Convert the baseline control quantity and target feedforward compensation quantity of the flying robot into motor speed commands, and control the flying robot to execute the motor speed commands.

[0073] In this embodiment of the invention, step 105 controls the flying robot to execute motor speed commands to adjust the flight parameters of the flying robot.

[0074] In this embodiment of the invention, the baseline control quantity is used to correct the trajectory and / or attitude of the flyable robot. Specifically, this baseline control quantity is used to achieve basic trajectory and attitude correction for the flyable robot. After incorporating the target feedforward compensation quantity, active suppression of wind resistance can be achieved, thereby realizing more refined trajectory and attitude correction. For example, after calculating the target feedforward compensation quantity and outputting the motor speed command, the flyable robot can be controlled to have an attitude angle deviation of <3° under level 5 wind conditions.

[0075] In this embodiment of the invention, by executing step 105, feedforward-feedback composite control of wind disturbance is realized, which not only maintains the effectiveness of the original trajectory and attitude correction functions, but also increases the active suppression capability of wind disturbance, which is beneficial to improving the flight stability and wind resistance performance of the robot when working at high altitudes.

[0076] It is evident that implementation Figure 2 The described wind-resistant control method for a flyable robot firstly acquires multi-dimensional sensor data in real time and combines it with state prediction equations to accurately separate the deviation caused by external wind disturbances, thus improving the accuracy of wind disturbance identification. Secondly, an adaptive law is used to estimate the prediction error, generating a wind disturbance torque that matches the current wind disturbance in real time, improving the robot's adaptability to sudden changes in wind speed and direction. Thirdly, by performing filtering and feedforward compensation calculations on the wind disturbance torque, the compensation amount is transformed into a precise target feedforward compensation amount, effectively suppressing the interference of measurement noise and estimation errors, and improving the smoothness and stability of the subsequently generated motor speed commands. Finally, the target feedforward compensation amount is fused with the baseline control amount and converted into a motor speed command, realizing feedforward-feedback composite control of wind disturbances. This maintains the effectiveness of the original trajectory and attitude correction functions while increasing the active suppression capability of wind disturbances, which is beneficial to improving the robot's flight stability and wind resistance performance during high-altitude operations.

[0077] In an optional embodiment, the method of performing filtering and feedforward compensation calculations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment specifically includes: The wind disturbance torque is subjected to low-pass filtering operation according to the preset low-pass filter to obtain the low-pass filtering result corresponding to the wind disturbance torque; Based on the preset feedforward compensation calculation formula and the preset rate saturation constraint, the target feedforward compensation amount for the low-pass filtering result is calculated; wherein, the rate saturation constraint is used to limit the rate of change of the target feedforward compensation amount.

[0078] In this optional embodiment, the purpose of setting up the low-pass filter (L1 filter) is to achieve decoupling between control and estimation, ensuring the continuity and feasibility of the control input. The transfer function of the low-pass filter can be:

[0079] in, Let be the cutoff frequency, s be the input signal, and the input of this low-pass filter be the estimated matching uncertainty, corresponding to the subsequent... The output is the L1 adaptive control input, and the calculation formula for this L1 adaptive control input is:

[0080] In this alternative embodiment, the signal is identified using the Laplace transform domain and satisfies C(0) = 1 to ensure that the DC gain is 1.

[0081] In this optional embodiment, in the discrete-time implementation, the recursive formula for the L1 low-pass filter is:

[0082] in, For intermediate variables of the filter, This is the final L1 adaptive control output.

[0083] In this optional embodiment, by performing a low-pass filtering operation on the wind disturbance torque through a preset low-pass filter, the high-frequency noise components mixed in the wind disturbance torque can be effectively filtered out, avoiding the control jitter problem caused by using noise signals as an effective compensation basis.

[0084] In this optional embodiment, the feedforward compensation calculation formula corresponding to the above feedforward compensation algorithm is as follows:

[0085] in, This is the feedforward compensation amount for the target; This is the feedforward gain matrix; the negative sign indicates that the compensation direction of the target feedforward compensation is opposite to the wind resistance direction. This is the wind disturbance moment.

[0086] In this optional embodiment, the feedforward compensation amount is used to represent the additional control amount that needs to be superimposed on the feedback control amount to compensate for system dynamic characteristics or external disturbances. In applications of flyable robots (quadrotor drones), the feedforward compensation amount Δu is typically related to the motor speed, and its magnitude and rate of change directly affect the motor's operating state and the system's control performance.

[0087] In this optional embodiment, the introduced rate saturation constraint takes into account the physical response limits of actuators such as motors, effectively preventing control mismatch or actuator damage caused by excessively rapid changes in compensation commands exceeding the actuator's response capability. Simultaneously, the rate saturation constraint can also effectively suppress sudden changes in compensation amounts that may be caused by abrupt wind disturbances, thereby improving the smoothness and stability of the compensation process.

[0088] In this optional embodiment, to prevent the feedforward compensation amount Δu from changing too rapidly, a corresponding rate saturation constraint mechanism / condition needs to be established. Let the feedforward compensation amount be Δu(t), and its rate of change should satisfy the following constraints:

[0089] Among them, v max This represents the maximum permissible rate of change.

[0090] In this optional embodiment, the rate saturation constraint in a discrete-time system can be expressed as:

[0091] To achieve rate limiting of the feedforward compensation amount Δu, rate limiting methods based on first-order low-pass filters, rate limiting methods based on dedicated rate limiters, or rate limiting methods based on extended state observers (ESOs) can be used, and the embodiments of the present invention are not limited thereto.

[0092] In this optional embodiment, furthermore, since there is a direct relationship between the feedforward compensation amount Δu and the motor speed, according to the dynamic model of the quadcopter UAV, the relationship between the motor speed and the thrust is T. i =kω i 2 Therefore, the feedforward compensation amount Δu is usually proportional to the square of the motor speed.

[0093] In this optional embodiment, the following method can be used to achieve rate saturation constraint on the motor speed: Firstly, the motor speed is limited, and the corresponding limiting formula is:

[0094] in, For the limited motor speed, For the desired motor speed, These are the lower and upper limits of the motor speed, respectively, typically ω. min =0 rpm, ω max =10000 rpm.

[0095] Secondly, regarding the limitation on the rate of change of rotational speed, the corresponding limitation formula is:

[0096] in, This represents the maximum permissible rate of change of the motor speed. The maximum permissible variation within each control cycle is related to the physical characteristics of the motor and the control performance requirements.

[0097] Third, a comprehensive rate saturation constraint algorithm is adopted, specifically: By combining the rate limit of the feedforward compensation amount Δu and the rate limit of the motor speed, the following integrated control algorithm can be designed: Calculate the expected feedforward compensation amount ; right Rate limiting is applied to obtain ; according to Calculate the desired motor speed ; right Amplitude and speed limits are applied to obtain the final motor speed command. ; The motor speed command is output to the motor driver.

[0098] As can be seen, in this optional embodiment, wind disturbance measurement noise is removed by low-pass filtering, and the rate of change is limited by rate saturation constraint. Accurate disturbance rejection is achieved by using directional feedforward compensation formula. Finally, a clean, smooth and physically feasible target feedforward compensation amount is output. This improves the calculation accuracy of the target feedforward compensation amount. Based on this accurate target feedforward compensation amount, it is beneficial to improve the stability, accuracy and execution reliability of subsequent wind resistance control of the robot.

[0099] Example 2 Please see Figure 3 , Figure 3 This is a flowchart illustrating another wind-resistant control method for a flyable robot disclosed in an embodiment of the present invention. Figure 3 The described wind-resistant control method for flying robots can be applied to wind-resistant control devices / systems for flying robots, and the embodiments of this invention are not limited thereto. Figure 3 As shown, the wind resistance control method for this flyable robot may include the following operations: 201. Construct the dynamic equations for the flyable robot. The dynamic equations are described using a six-degree-of-freedom system, which includes three translational degrees of freedom and three rotational degrees of freedom.

[0100] In this embodiment of the invention, the six degrees of freedom description includes translational motion in the forward, backward, left, right, and up / down directions, as well as roll, pitch, and yaw attitude changes. This refined degree of freedom description provides a sufficient data foundation for subsequent wind resistance control, helping to avoid wind disturbance estimation errors caused by incomplete motion descriptions.

[0101] In this embodiment of the invention, in the inertial coordinate system {A}, the position vector of the body's center of mass is denoted as r=(x,y,z) ∈A, and the attitude of the body coordinate system {B} relative to {A} is determined by the rotation matrix: It means that, among them, Let be a three-dimensional rotation group, which is the set of all orthogonal matrices with a determinant of +1, used to represent all possible rotations of a rigid body in three-dimensional space. The x-coordinates of the machine coordinate system are respectively B Axis (usually pointing towards the machine head), y B Axis (usually pointing to the right wing), z B The direction vector of the axis (usually pointing downwards or upwards, depending on the definition) in {A}. The linear velocity of the body relative to the inertial frame is v ∈A, and the angular velocity is Ω∈B.

[0102] In this embodiment of the invention, the complete dynamic equations of the flyable robot (quadrotor drone) include translational motion equations and rotational motion equations, wherein the calculation formula corresponding to the translational motion equation is:

[0103]

[0104] The calculation formula corresponding to this rotational motion equation is:

[0105]

[0106] Where m is the robot's mass and g is the acceleration due to gravity. Let {A} be the unit vector along the z-axis of the inertial coordinate system. Let be the time derivative of the rotation matrix, used to describe the rate of change of attitude; R is the rotation matrix; I ∈ R 3×3 Let Ω be the inertial matrix of the aircraft, F∈B be the total thrust vector generated by the rotor, and τ∈B be the total torque vector generated by the rotor. × Let Ω be an antisymmetric matrix composed of angular velocity vectors Ω, satisfying Ω × v = Ω × For any vector v∈R 3 Established.

[0107] In this embodiment of the invention, the four rotors of the flying robot can be configured according to a specific rotation direction. For example, even-numbered rotors rotate counterclockwise, and odd-numbered rotors rotate clockwise. Optionally, the steady-state thrust generated by each rotor can be modeled as follows:

[0108] in, For thrust coefficient, air density, For rotor sweep area, The rotor radius is... This represents the rotor angular velocity.

[0109] In this embodiment of the invention, the thrust and torque of the four rotors are combined to obtain the total force and torque vector in the airframe coordinate system. Let the distance from the rotor to the center of the airframe be... l Then the total thrust and torque can be expressed as:

[0110]

[0111] in, F Let {B} be the total thrust vector in the body coordinate system; This is the total moment vector in the body coordinate system {B}; denoted as , where is the thrust of the i-th rotor; k is the anti-torque coefficient, which is related to the aerodynamic characteristics of the rotor.

[0112] 202. Convert the dynamic equations into state prediction equations and initialize the state prediction equations. The state prediction equations include multiple state vectors and multiple control input vectors.

[0113] In this embodiment of the invention, the plurality of state vectors include at least a first sub-vector corresponding to the position, velocity, attitude angle and angular velocity of the aircraft's center of mass; the plurality of control input vectors include at least a second sub-vector corresponding to the total thrust, roll torque, pitch torque and yaw torque.

[0114] In this embodiment of the invention, in order to adapt to the design of the L1 adaptive controller, the dynamic equation needs to be converted into a state-space form.

[0115] In this embodiment of the invention, the state vector of the control system corresponding to the flying robot is defined as follows: ,in, Let Z be the attitude angle vector (represented using ZXY Euler angles). Let the control input vector be: ,in, For total thrust, These are the roll, pitch, and yaw moments, respectively.

[0116] In this embodiment of the invention, the kinetic equation constructed in step 201 above is:

[0117] in, ,and This describes the system status of the flying robot. Baseline control input; For adaptive control input (corresponding to the output of L1 adaptive controller); To match uncertainty, This represents non-matching uncertainty; , These are the uncertainty distribution matrices for matched and unmatched products, respectively.

[0118] In this embodiment of the invention, the state prediction equation transformed in step 202 is:

[0119] in, ,and The state corresponding to the preset predictor (here we only consider translation and rotation states that are directly affected by uncertainty). To address this uncertainty The estimate; , This represents the prediction error; , and is a Hurwitz matrix; For the augmented distribution matrix, and .

[0120] 203. Real-time acquisition of sensor detection data for flying robots.

[0121] 204. Based on the sensor detection data and the pre-constructed state prediction equation, the prediction error of the flying robot is calculated.

[0122] 205. Perform data estimation on the prediction error according to the preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error.

[0123] 206. Perform filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment.

[0124] 207. Convert the baseline control quantity and target feedforward compensation quantity of the flying robot into motor speed commands, and control the flying robot to execute the motor speed commands.

[0125] For further descriptions of steps 203-207 in this embodiment of the invention, please refer to the other specific descriptions of steps 101-105 in Embodiment 1. These descriptions will not be repeated in this embodiment of the invention.

[0126] It is evident that implementation Figure 3 The wind-resistant control method for the described flyable robot establishes a complete motion state description system and structured parameter configuration by constructing a six-degree-of-freedom dynamic equation and converting it into a state prediction equation. This provides a precise error calculation benchmark and physical model support for subsequent wind-resistant control, which is beneficial to improving the accuracy of subsequent state prediction.

[0127] In an optional embodiment, step 204 above, which calculates the prediction error of the flyable robot based on sensor detection data and a pre-constructed state prediction equation, specifically includes the following methods: The difference between the state prediction equation and the dynamic equation is calculated to obtain the prediction error dynamic equation; According to the preset fusion algorithm, a data fusion operation is performed on all sub-detection data included in the sensing detection data to obtain a data fusion result corresponding to all sub-detection data; the preset fusion algorithm includes a fusion algorithm based on Kalman filter; the data fusion result is a multi-dimensional fusion vector corresponding to all state vectors; Substituting the data fusion results into the prediction error dynamic equation yields the prediction error of the flying robot.

[0128] In this optional embodiment, by calculating the difference between the state prediction equation and the dynamic equation, a prediction error dynamic equation is constructed. This can transform the originally implicit prediction error into a dynamic system with a clear mathematical form, so that the change law of the error can be tracked and analyzed in real time, providing a clear, explicit and specific error evolution model for subsequent adaptive law estimation.

[0129] In this optional embodiment, the dynamic equation for the prediction error calculated above is:

[0130] in, And it is a generalized uncertainty term, in which, To match uncertainty.

[0131] As can be seen, in this optional embodiment, an explicit dynamic representation of the error is achieved by constructing a prediction error dynamic equation, and high-precision fusion of multi-source data is achieved by combining the Kalman filter fusion algorithm. Furthermore, the Hurwitz matrix is ​​used to ensure asymptotic convergence of the error, and finally, an accurate prediction error dynamic equation is constructed. Based on this prediction error dynamic equation, the estimation accuracy of wind disturbance is greatly improved.

[0132] Example 3 Please see Figure 4 , Figure 4 This is a schematic diagram of the wind-resistant control system for a flyable robot disclosed in an embodiment of the present invention. The wind-resistant control system of this flyable robot can be used to execute some or all of the steps in any wind-resistant control method for a flyable robot described in Embodiment 1 or Embodiment 2 of the present invention; the embodiments of the present invention are not limited thereto. Figure 4 As shown, the wind-resistant control system of the flyable robot may include a data acquisition module 301, an error calculation module 302, a data estimation module 303, a feedforward compensation calculation module 304, and a command conversion and execution module 305, wherein: The data acquisition module 301 is used to acquire sensor detection data for the flying robot in real time.

[0133] The error calculation module 302 is used to calculate the prediction error of the flying robot based on the sensor detection data and the pre-constructed state prediction equation; the state prediction equation is used to predict the flight data of the flying robot in an ideal state; the sensor detection data includes at least sub-detection data corresponding to multiple preset state vectors.

[0134] The data estimation module 303 is used to perform data estimation on the prediction error according to the preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error.

[0135] The feedforward compensation calculation module 304 is used to perform filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment.

[0136] The instruction conversion and execution module 305 is used to convert the baseline control quantity and target feedforward compensation quantity of the flying robot into motor speed commands, and control the flying robot to execute the motor speed commands to adjust the flight parameters of the flying robot; the baseline control quantity is used to correct the trajectory and / or attitude of the flying robot.

[0137] It is evident that implementation Figure 4The wind-resistant control system for the described flyable robot firstly acquires multi-dimensional sensor data in real time and combines it with state prediction equations to accurately separate the deviation caused by external wind disturbances, thus improving the accuracy of wind disturbance identification. Secondly, by using an adaptive law to estimate the prediction error, it can generate a wind disturbance torque that matches the current wind disturbance in real time, improving the robot's adaptability to sudden changes in wind speed and direction. Thirdly, by performing filtering and feedforward compensation calculations on the wind disturbance torque, the compensation amount is transformed into a precise target feedforward compensation amount, effectively suppressing the interference of measurement noise and estimation errors, and improving the smoothness and stability of the subsequently generated motor speed commands. Finally, the target feedforward compensation amount is fused with the baseline control amount and converted into a motor speed command, realizing feedforward-feedback composite control of wind disturbances. This maintains the effectiveness of the original trajectory and attitude correction functions while increasing the active suppression capability of wind disturbances, which is beneficial to improving the robot's flight stability and wind resistance performance during high-altitude operations.

[0138] In an optional embodiment, please refer to Figure 5 , Figure 5 This is a schematic diagram of the wind-resistant control system for a flyable robot disclosed in an embodiment of the present invention. Figure 5 As shown, the system may also include a construction module 306 and an equation transformation module 307, wherein: Module 306 is used to construct the dynamic equations for the flyable robot, which are described by a six-degree-of-freedom system including three translational degrees of freedom and three rotational degrees of freedom. Equation transformation module 307 is used to convert dynamic equations into state prediction equations and initialize state prediction equations. The state prediction equations include multiple state vectors and multiple control input vectors. The multiple state vectors include at least a first sub-vector corresponding to the position, velocity, attitude angle and angular velocity of the body center of mass. The multiple control input vectors include at least a second sub-vector corresponding to the total thrust, roll moment, pitch moment and yaw moment.

[0139] As can be seen, in this optional embodiment, by constructing a six-degree-of-freedom dynamic equation and converting it into a state prediction equation, a complete motion state description system and structured parameter configuration are established, providing an accurate error calculation benchmark and physical model support for subsequent wind resistance control, which is conducive to improving the accuracy of subsequent state prediction.

[0140] In another optional embodiment, the error calculation module 302 calculates the prediction error of the flyable robot based on the sensor detection data and the pre-built state prediction equation in the following specific ways: The difference between the state prediction equation and the dynamic equation is calculated to obtain the prediction error dynamic equation; According to the preset fusion algorithm, a data fusion operation is performed on all sub-detection data included in the sensing detection data to obtain a data fusion result corresponding to all sub-detection data; the preset fusion algorithm includes a fusion algorithm based on Kalman filter; the data fusion result is a multi-dimensional fusion vector corresponding to all state vectors; Substituting the data fusion results into the prediction error dynamic equation yields the prediction error of the flying robot.

[0141] In this optional embodiment, the above-mentioned kinetic equations are:

[0142] in, ,and This describes the system status of the flying robot. Baseline control input; For adaptive control input; To match uncertainty, This represents non-matching uncertainty; These are the uncertainty distribution matrices for matched and unmatched products, respectively; The state prediction equation is:

[0143] in, ,and The state corresponding to the preset predictor (here we only consider translation and rotation states that are directly affected by uncertainty). To address this uncertainty The estimate; , This represents the prediction error; , and is a Hurwitz matrix; For the augmented distribution matrix, and .

[0144] In this optional embodiment, the prediction error dynamic equation is:

[0145] in, And it is a generalized uncertainty term, in which, To match uncertainty.

[0146] As can be seen, in this optional embodiment, an explicit dynamic representation of the error is achieved by constructing a prediction error dynamic equation, and high-precision fusion of multi-source data is achieved by combining the Kalman filter fusion algorithm. Furthermore, the Hurwitz matrix is ​​used to ensure asymptotic convergence of the error, and finally, an accurate prediction error dynamic equation is constructed. Based on this prediction error dynamic equation, the estimation accuracy of wind disturbance is greatly improved.

[0147] In another optional embodiment, the feedforward compensation calculation module 304 performs filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment, specifically including: The wind disturbance torque is subjected to low-pass filtering operation according to the preset low-pass filter to obtain the low-pass filtering result corresponding to the wind disturbance torque; Based on the preset feedforward compensation calculation formula and the preset rate saturation constraint, the target feedforward compensation amount for the low-pass filtering result is calculated; wherein, the rate saturation constraint is used to limit the rate of change of the target feedforward compensation amount.

[0148] In this optional embodiment, the feedforward compensation calculation formula is:

[0149] in, This is the feedforward compensation amount for the target; This is the feedforward gain matrix; the negative sign indicates that the compensation direction of the target feedforward compensation is opposite to the wind resistance direction. This is the wind disturbance moment.

[0150] As can be seen, in this optional embodiment, wind disturbance measurement noise is removed by low-pass filtering, and the rate of change is limited by rate saturation constraint. Accurate disturbance rejection is achieved by using directional feedforward compensation formula. Finally, a clean, smooth and physically feasible target feedforward compensation amount is output. This improves the calculation accuracy of the target feedforward compensation amount. Based on this accurate target feedforward compensation amount, it is beneficial to improve the stability, accuracy and execution reliability of subsequent wind resistance control of the robot.

[0151] Example 4 Please see Figure 6 , Figure 6 This is a structural schematic diagram of a wind-resistant control device for a flying robot disclosed in an embodiment of the present invention. Figure 6 As shown, the wind-resistant control device of the flying robot may include: Memory 401 storing executable program code; Processor 402 coupled to memory 401; The processor 402 calls the executable program code stored in the memory 401 to execute some or all of the steps in the wind-resistant control method for any of the flying robots described in Embodiment 1 or Embodiment 2 of the present invention.

[0152] Example 5 This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute some or all of the steps in the wind-resistant control method for any of the flying robots described in Embodiment 1 or Embodiment 2 of this invention.

[0153] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0154] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0155] Finally, it should be noted that the above embodiments are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit them. Although the present invention 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 the present invention.

Claims

1. A wind-resistant control method for a flyable robot, characterized in that, The method includes: The system acquires real-time sensor detection data for the flyable robot and calculates the prediction error of the flyable robot based on the sensor detection data and a pre-constructed state prediction equation. The state prediction equation is used to predict the flight data of the flyable robot in an ideal state. The sensor detection data includes at least sub-detection data corresponding to multiple preset state vectors. The prediction error is estimated according to a preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error; and the wind disturbance moment is filtered and fedforward compensation is calculated to obtain the target feedforward compensation amount corresponding to the wind disturbance moment. The baseline control quantity of the flyable robot and the target feedforward compensation quantity are converted into motor speed commands, and the flyable robot is controlled to execute the motor speed commands to adjust the flight parameters of the flyable robot; the baseline control quantity is used to correct the trajectory and / or attitude of the flyable robot.

2. The wind-resistant control method for a flyable robot according to claim 1, characterized in that, The method further includes: A dynamic equation is constructed for a flyable robot, which is described by a six-degree-of-freedom system including three translational degrees of freedom and three rotational degrees of freedom. The dynamic equations are converted into state prediction equations and initialized. The state prediction equations include multiple state vectors and multiple control input vectors. The multiple state vectors include at least a first sub-vector corresponding to the body's center of mass position, velocity, attitude angle, and angular velocity. The multiple control input vectors include at least a second sub-vector corresponding to the total thrust, roll moment, pitch moment, and yaw moment.

3. The wind resistance control method for a flyable robot according to claim 2, characterized in that, The calculation of the prediction error of the flyable robot based on the sensor detection data and the pre-constructed state prediction equation includes: The difference between the state prediction equation and the dynamic equation is calculated to obtain the prediction error dynamic equation; According to a preset fusion algorithm, a data fusion operation is performed on all the sub-detection data included in the sensing detection data to obtain a data fusion result corresponding to all the sub-detection data; the preset fusion algorithm includes a fusion algorithm based on a Kalman filter; the data fusion result is a multi-dimensional fusion vector corresponding to all the state vectors; Substituting the data fusion result into the prediction error dynamic equation yields the prediction error of the flyable robot.

4. The wind-resistant control method for a flyable robot according to any one of claims 1-3, characterized in that, The step of performing filtering and feedforward compensation calculations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment includes: A low-pass filtering operation is performed on the wind disturbance torque according to a preset low-pass filter to obtain a low-pass filtering result corresponding to the wind disturbance torque; Based on the preset feedforward compensation calculation formula and combined with the preset rate saturation constraint, the target feedforward compensation amount for the low-pass filtering result is calculated; wherein, the rate saturation constraint is used to limit the rate of change of the target feedforward compensation amount.

5. The wind-resistant control method for a flyable robot according to claim 4, characterized in that, The formula for calculating the feedforward compensation is as follows: in, This is the feedforward compensation amount for the target; This is the feedforward gain matrix; the negative sign indicates that the direction of the feedforward compensation for this target is opposite to the direction of wind resistance. This is the wind disturbance moment.

6. The wind-resistant control method for a flying robot according to claim 3, characterized in that, The dynamic equation is: in, ,and This describes the system status of the flying robot. Baseline control input; For adaptive control input; To match uncertainty, This represents non-matching uncertainty; These are the uncertainty distribution matrices for matched and unmatched products, respectively; The state prediction equation is: in, ,and The state corresponding to the preset predictor. To address this uncertainty The estimate; , This represents the prediction error; , and is a Hurwitz matrix; For the augmented distribution matrix, and .

7. The wind-resistant control method for a flyable robot according to claim 6, characterized in that, The dynamic equation for the prediction error is: in, And it is a generalized uncertainty term, in which, To match uncertainty.

8. A wind-resistant control system for a flying robot, characterized in that, The system includes: The data acquisition module is used to acquire sensor detection data for the flying robot in real time; An error calculation module is used to calculate the prediction error of the flyable robot based on the sensor detection data and a pre-constructed state prediction equation; the state prediction equation is used to predict the flight data of the flyable robot in an ideal state; the sensor detection data includes at least sub-detection data corresponding to multiple preset state vectors; The data estimation module is used to perform data estimation on the prediction error according to a preset adaptive law to obtain the wind disturbance moment corresponding to the prediction error; The feedforward compensation calculation module is used to perform filtering and feedforward compensation calculation operations on the wind disturbance moment to obtain the target feedforward compensation amount corresponding to the wind disturbance moment. The instruction conversion and execution module is used to convert the baseline control quantity and the target feedforward compensation quantity of the flyable robot into motor speed commands, and control the flyable robot to execute the motor speed commands to adjust the flight parameters of the flyable robot; the baseline control quantity is used to correct the trajectory and / or attitude of the flyable robot.

9. A wind-resistant control device for a flying robot, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the wind-resistant control method for the flyable robot as described in any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the wind-resistant control method for the flyable robot as described in any one of claims 1-7.

Images