A method and system for unmanned aerial vehicle attitude stabilization control oriented to special environmental disturbance

Abstract

This invention belongs to the field of unmanned aerial vehicle (UAV) control technology, specifically relating to a method and system for attitude stabilization control of UAVs under special environmental disturbances. The method includes the following steps: S1: Establishing a wind field disturbance model, a dust particle impact model, and an electromagnetic interference model, and dynamically coupling these three models into a total disturbance vector experienced by the UAV; S2: Constructing a six-degree-of-freedom dynamic model incorporating the influence of the total disturbance vector. This model describes the translational and rotational motion of the UAV's center of mass under multi-source disturbances in complex environments; S3: Based on traditional aerodynamic models and incorporating environmental disturbance parameters, constructing a variable structure model with aerodynamic coefficients. This invention effectively improves the UAV's adaptability to multi-source coupled disturbances by constructing a unified multi-layered architecture for environmental identification, disturbance observation, Lyapunov robust control, and actuator compensation, thereby ensuring mission accuracy and safety.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle control technology, specifically relating to a method and system for attitude stabilization control of unmanned aerial vehicles facing special environmental disturbances. Background Technology

[0002] With the surge in the number of missions in complex environments, drones have become key equipment in fields such as power line inspection, disaster detection, and marine monitoring. However, under harsh conditions such as strong winds, sandstorms, and electromagnetic interference, conventional drones struggle to maintain stable attitudes, easily experiencing severe attitude fluctuations, trajectory deviations, or even instability and crashes, directly threatening mission safety and coverage efficiency. Existing control methods are mostly based on ideal environment assumptions, relying solely on linear models and simple compensation strategies. Once encountering wind gusts, particle impacts, or high-frequency electromagnetic noise, the controller response is slow and compensation is insufficient, leading to a rapid accumulation of attitude errors.

[0003] To mitigate risks, operators are forced to shorten mission durations, increase manual monitoring, or install additional protective devices, which not only raises maintenance costs but also increases the risk of secondary accidents. While some drones have reinforced sensors, they lack a unified framework for environmental identification and disturbance compensation, failing to correct their attitude in real time under multi-source coupled disturbances. This demonstrates that existing technologies are significantly inadequate in adapting to special environments, necessitating a systematic solution that integrates environmental modeling, disturbance observation, robust control, and actuator compensation to ensure drones maintain stable attitudes even in harsh conditions, guaranteeing mission continuity and flight safety. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for attitude stabilization control of unmanned aerial vehicles (UAVs) in response to special environmental disturbances. By constructing a unified multi-layer architecture for environmental identification, disturbance observation, Lyapunov robust control, and actuator compensation, the invention effectively improves the UAV's adaptability to multi-source coupled disturbances, thereby ensuring mission accuracy and safety.

[0005] The specific technical solution adopted by this invention is as follows: A method for attitude stabilization control of unmanned aerial vehicles (UAVs) under special environmental disturbances includes the following steps: S1: Establish wind field disturbance model, dust particle impact model and electromagnetic interference model, and dynamically couple the three into the total disturbance vector experienced by the UAV; The total disturbance vector is constructed by integrating the total disturbance force and the total disturbance torque that change dynamically with time; The total disturbance force is composed of the vector sum of the aerodynamic force and the impact force, and the total disturbance torque is composed of the vector sum of the aerodynamic torque, the impact torque, and the equivalent torque. Based on the real-time wind speed, particulate matter concentration, and electromagnetic field intensity data collected by the environmental information acquisition unit, the gust parameters of the wind field disturbance model, the particle distribution parameters of the dust particle impact model, and the noise parameters of the electromagnetic interference model are corrected online. S2: Construct a six-degree-of-freedom dynamic model that includes the influence of the total disturbance vector. The model is used to describe the translational motion of the center of mass and the rotational motion around the center of mass of the UAV under multi-source disturbances in a complex environment. The six-degree-of-freedom dynamic model integrates the UAV's mass, center-of-mass velocity vector, gravitational acceleration vector, rotation matrix updated in real time by attitude quaternions, propulsion aerodynamic force, wind field additional force, and dust particle impact force. The inertia matrix, the body angular velocity vector, as well as the aerodynamic torque, wind torque, particle impact torque, and electromagnetic interference torque, and the attitude quaternion are updated in real time with the evolution of angular velocity to ensure dynamic matching between dynamic characteristics and environmental disturbances. S3: Based on the traditional aerodynamic model, and by introducing environmental disturbance parameters, construct an aerodynamic coefficient variable structure model; use the environmental disturbance parameters in step S1 as variables of aerodynamic coefficients, and dynamically adjust the lift coefficient, pitching moment coefficient, and yaw moment coefficient. In the model, the angle of attack gain, sideslip gain and environmental disturbance parameters are linked, the aerodynamic damping derivative is corrected in real time by the coupling change of the body angular velocity and environmental disturbance, and the dynamic pressure, airspeed and wind field disturbance are matched in real time to enable the aerodynamic characteristics to adapt to special environments. S4: Based on the model in steps S2 and S3, the Lyapunov function is constructed by using the error state vector and analyzing the convergence of attitude error, and its time derivative is derived to obtain the parameter constraints that make the system stable or uniformly ultimately bounded under multi-source bounded perturbations. S5: Under the constraints of step S4, the output of the attitude control law is combined with the disturbance estimate of the observer, and the multiplicative and additive electromagnetic interference in the actuator is corrected by the adaptive inverse compensation module to generate the actual command to drive the actuator. The attitude control law design adopts a composite structure of proportional term, derivative term and environmental compensation term. The proportional term and derivative term correspond to the real-time correction of attitude error and angular velocity error, respectively. The environmental compensation term directly calls the real-time output of the disturbance model in step S1 to offset the disturbance effect in advance. Observer design: A sliding mode observer architecture is adopted. By designing a nonlinear sliding surface and an adaptive sliding mode gain, finite-time convergence estimation of wind field abrupt changes, dust impacts, and electromagnetic interference is achieved. The adaptive inverse compensation module addresses command distortion caused by electromagnetic interference by constructing an inverse model of the actuator input based on the multiplicative disturbance estimate and additive noise estimate from the observer output. Through identity matrix normalization, it corrects the control command in real time. S6. Linearize the nonlinear closed-loop control system formed in step S5 near the desired equilibrium point to obtain a linear state-space model; by analyzing the distribution of the closed-loop system poles when the key environment and system parameters change, determine whether it meets the preset stability margin, thereby determining the stable operating region of the system; if the poles exceed the stable region, adjust the feedback gain matrix to reconfigure the poles.

[0006] It also includes collecting attitude data and environmental parameters in real time through sensors on the drone, updating the disturbance model of step S1, the dynamic model of step S2 and the aerodynamic coefficient model of step S3 every 10ms, and simultaneously correcting the control law parameters of step S5 and the stability domain boundary of step S6, forming a closed-loop iteration of perception-modeling-control-evaluation.

[0007] An attitude stabilization control system for unmanned aerial vehicles (UAVs) accommodating special environmental disturbances includes: The attitude measurement unit is used to collect inertial measurement information such as attitude quaternions and angular velocity of the UAV, providing input for attitude error calculation and dynamic state update; The environmental information acquisition unit is used to acquire or estimate wind field disturbance parameters, dust particle concentration, electromagnetic interference intensity, and other environmental disturbance information that affects aerodynamic changes. It provides a basis for disturbance modeling and parameter estimation by continuously collecting environmental data. The flight control processing unit is used to perform attitude error calculation, disturbance observer update, attitude control law solution, and control command generation in real time based on the data provided by the attitude measurement unit and environmental information acquisition unit during the flight of the UAV. Based on the pre-built disturbance environment model, six-degree-of-freedom dynamic model, and aerodynamic coefficient variable parameter model, it calls the feedback gain matrix and pole placement analysis results obtained from offline calculation to achieve compensation for multi-source disturbances and attitude stability control. The actuator drive unit is used to drive the actions of each actuator according to the attitude control commands output by the flight control processing unit, and, in conjunction with the output of the adaptive inverse compensation module, corrects the input distortion caused by electromagnetic interference or dynamic deviation of the actuator to ensure that the control commands are executed accurately.

[0008] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the above-described method.

[0009] The technical effects achieved by this invention are as follows: This invention constructs a unified environmental model for multiple sources of disturbance, including wind field disturbances, dust particle impacts, and electromagnetic interference, and combines this with variable parameter modeling of aerodynamic coefficients, enabling attitude dynamics to accurately reflect the characteristics of complex environments. The control method based on Lyapunov stability analysis and disturbance observer design can estimate and compensate for external disturbances in real time, ensuring that attitude errors converge quickly even in harsh environments. Through pole placement stability domain analysis, this invention further clarifies the stability margin of the attitude control system under parameter variations, improving the reliability of the controller in multi-source disturbance scenarios. Attached Figure Description

[0010] Figure 1 This is a flowchart of a method for attitude stabilization control of a UAV for special environments according to the present invention; Figure 2 This is a schematic diagram of the system structure of an attitude stabilization control system for unmanned aerial vehicles (UAVs) designed for special environments, according to the present invention. Figure 3 This is a schematic diagram of the disturbance environment modeling of an attitude stabilization control method and system for unmanned aerial vehicles (UAVs) in special environments according to the present invention. Figure 4 This is a schematic diagram of the principle of an attitude stabilization control method for unmanned aerial vehicles (UAVs) designed for special environments, according to the present invention. Detailed Implementation

[0011] To make the objectives and advantages of this invention clearer, the invention will be specifically described below with reference to embodiments. It should be understood that the following text is merely used to describe one or more specific embodiments of the invention and does not strictly limit the scope of protection specifically claimed by the invention.

[0012] like Figure 1 , Figure 3 and Figure 4 As shown, a method for attitude stabilization control of a UAV under special environmental disturbances includes the following steps: S1: Establish wind field disturbance model, dust particle impact model and electromagnetic interference model, and dynamically couple the three into the total disturbance vector experienced by the UAV; The wind field disturbance model is constructed based on the principle of superposition of mean wind and random gusts, and the wind speed field is represented as: in Let be the wind speed vector, representing the three-dimensional wind speed at time t and position r. The average wind speed distribution is denoted as . ; Let be the amplitude vector of the i-th gust. As the central location, For spatial scale parameters, Angular frequency, Let N be the initial phase and N be the number of gusts. The gust term is characterized by a Gaussian spatial distribution and a sinusoidal time variation. To ensure that random gusts have statistical characteristics consistent with actual measurements, the Kaimal wind spectrum is introduced to constrain the gust parameters. Its longitudinal wind speed spectrum can be expressed as: in The standard deviation of wind speed along the principal direction, For turbulence integral scale, The average wind speed scalar. For frequency; the Kaimal spectrum is used to generate gust inputs consistent with measured statistics online; When a drone traverses a dusty area, considering the impact force of the particulate gas-solid two-phase flow on the drone's surface, a dust particle impact model is established: in To record the particle density distribution, For particle velocity, Local air velocity of the machine body The particle drag coefficient, The effective impact area characterizes the effect of a single particle; In a strong electromagnetic environment, the actuator drive signal will be superimposed with multiplicative disturbances and additive biases. The actual output of the actuator can be modeled as follows: in For control commands, To ensure actual output, For multiplicative perturbations, This is additive noise; through the unified construction of the above three types of disturbance models, an overall environmental disturbance model suitable for subsequent control design is obtained; The total disturbance vector is constructed by integrating the total disturbance force and the total disturbance torque that change dynamically with time; The total disturbance force is composed of the vector sum of the aerodynamic force and the impact force, and the total disturbance torque is composed of the vector sum of the aerodynamic torque, the impact torque, and the equivalent torque. The additional aerodynamic forces and aerodynamic moments caused by the wind field are denoted as... , The impact of sand and dust particles is equivalent to the impact force and impact torque. , The effect of electromagnetic interference on the actuator channel is equivalent to the gain perturbation and additional noise of the control input, and its equivalent moment is denoted as... The total disturbance force can be obtained by superposition. and disturbance torque : , This allows multi-source disturbances in complex environments to be uniformly represented as time-varying disturbance vectors. This provides a clear disturbance path for subsequent dynamic modeling and robust control design; Based on the real-time wind speed, particulate matter concentration, and electromagnetic field intensity data collected by the environmental information acquisition unit, the gust parameters of the wind field disturbance model, the particle distribution parameters of the dust particle impact model, and the noise parameters of the electromagnetic interference model are corrected online. S2: To describe the rigid body motion of the UAV under complex disturbances, a six-degree-of-freedom dynamic model is established to describe the coupled behavior of the center-of-mass translation and attitude rotation; the specific model is as follows: in For the quality of drones, Let the velocity vector be the center of mass. The vector of gravitational acceleration. Represents the attitude quaternion Constructed rotation matrix, For propulsion aerodynamics under the machine system, Adding force to the wind field, The impact force of sand and dust particles, The inertia matrix, The vector of the body's angular velocity. , , These are aerodynamic torque, wind field torque, and particle impact torque, respectively. Additional torque caused by electromagnetic interference from the actuator; attitude quaternion The process of angular velocity evolution is updated in real time to satisfy the following: in This is an antisymmetric matrix related to angular velocity, providing a fundamental mathematical description for subsequent attitude control and stability analysis. S3: To reflect the impact of special environments on aerodynamic characteristics, the environmental disturbance parameters in step S1 are used as variables for aerodynamic coefficients, and the lift coefficient, pitching moment coefficient, and yaw moment coefficient are dynamically adjusted to construct a variable structure model of aerodynamic coefficients. The specific model is as follows: in , , As the reference aerodynamic parameters, , , These are the angle-of-attack gain, the influence coefficient of angle of attack on torque, and the sideslip angle gain. For the angle of attack, Sideslip angle, , , For parameters related to environmental disturbances, To characterize environmental disturbance parameters such as wind field intensity and particle concentration, , , The aerodynamic damping derivative is related to angular velocity. For dynamic pressure, The chord length, For airspeed; by organizing the above coefficients into a mapping matrix from control input to aerodynamic torque, the aerodynamic damping derivative is corrected in real time for the coupling change of random body angular velocity and environmental disturbance, and dynamic pressure, airspeed and wind field disturbance are matched in real time for adaptive adjustment of aerodynamic characteristics to environmental disturbance. S4: To analyze the stability of attitude error under multi-source disturbances, the error state vector is constructed: Where is the attitude error quaternion vector, is the body angular velocity error vector, and is the environmental parameter estimation error; To analyze the convergence of attitude error, a Lyapunov function is constructed: in and Given a positive definite gain matrix, based on the dynamic and aerodynamic models obtained in steps S2 and S3, the dynamic equations of the error system are derived, and the derivative of the Lyapunov function is calculated. Its upper bound expression is given. Under the condition of reasonably selecting the control gain and observer parameters, the following can be obtained: in It is a symmetric positive definite matrix. For equivalent external disturbance, Given bounded constants, we obtain the parameter constraints that make the system stable or uniformly ultimately bounded under multi-source bounded perturbations. S5: Under the constraints of step S4, the output of the attitude control law is combined with the disturbance estimate of the observer, and the multiplicative and additive electromagnetic interference in the actuator is corrected by the adaptive inverse compensation module to generate the actual command to drive the actuator. The attitude control law design adopts a composite structure of proportional, derivative, and environmental compensation terms. The proportional and derivative terms correspond to the real-time correction of attitude error and angular velocity error, respectively. The environmental compensation term directly calls the real-time output of the disturbance model in step S1 to offset the disturbance effect in advance. The control law design is as follows: in For nominal control input, This is the angular velocity feedback gain matrix. These are estimated environmental parameters, and these are estimated disturbance torques. To estimate disturbances in real time, a nonlinear sliding mode surface and adaptive sliding mode gain are designed to achieve finite-time convergent estimation of wind field abrupt changes, dust impacts, and electromagnetic interference. A sliding mode observer is designed as follows: in and For nonlinear functions determined by the model, For sliding mode gain, For symbolic functions, The sliding surface is related to the perturbation estimation error; it can be proven using the Lyapunov method that, with appropriate selection... At that time, it can converge to the vicinity of the real disturbance in a finite amount of time; To address input distortion caused by electromagnetic interference that may affect the actuator, an adaptive inverse compensation module is introduced based on the multiplicative perturbation estimate and additive noise estimate from the observer output. This module constructs the inverse model of the actuator input and generates actuator commands. in It is the identity matrix. This is the estimated value of the multiplicative disturbance. The additive noise estimate is normalized using an identity matrix, and the control command is corrected in real time. S6. To evaluate the stability region of the attitude control system under multi-source disturbances and parameter uncertainties, the nonlinear closed-loop system obtained in step S5 is linearized to the first order near the desired equilibrium attitude; the linearized state vector is selected as follows: Under the assumption of small perturbations, the linear state-space model of the closed-loop system can be obtained: in, To linearize the system matrix, The input matrix is ​​linearized, and the disturbance term is linearized; the feedback form of the feedback control law in step S5 is then linearized. Substituting the values, we obtain the closed-loop linear system as follows: in The linearized feedback gain matrix is ​​obtained by calculating the matrix. eigenvalues The drift of the closed-loop poles is analyzed under the disturbances of key parameters such as wind field intensity, dust particle concentration, electromagnetic interference amplitude, sensor noise, and actuator bandwidth variation; if all parameters are within the allowable range of variation, the following conditions are met: in To preset the stability margin threshold, the stability domain and allowable variation range of key parameters of the attitude control system under the aforementioned multi-source disturbance conditions can be determined accordingly. If the poles cross the stability half-plane boundary under certain operating conditions, the feedback gain matrix can be adjusted. The closed-loop poles are reconfigured to return to the region that meets the stability margin requirements, thereby ensuring that the UAV maintains sufficient stable flight capability in harsh environments; It also includes collecting attitude data and environmental parameters in real time through sensors on the drone, updating the disturbance model of step S1, the dynamic model of step S2 and the aerodynamic coefficient model of step S3 every 10ms, and simultaneously correcting the control law parameters of step S5 and the stability domain boundary of step S6, forming a closed-loop iteration of perception-modeling-control-evaluation.

[0013] like Figure 2 As shown, an attitude stabilization control system for unmanned aerial vehicles (UAVs) accommodating special environmental disturbances includes: The attitude measurement unit is used to collect inertial measurement information such as attitude quaternions and angular velocity of the UAV, providing input for attitude error calculation and dynamic state update; The environmental information acquisition unit is used to acquire or estimate wind field disturbance parameters, dust particle concentration, electromagnetic interference intensity, and other environmental disturbance information that affects aerodynamic changes. It provides a basis for disturbance modeling and parameter estimation by continuously collecting environmental data. The flight control processing unit is used to perform attitude error calculation, disturbance observer update, attitude control law solution, and control command generation in real time based on the data provided by the attitude measurement unit and environmental information acquisition unit during the flight of the UAV. Based on the pre-built disturbance environment model, six-degree-of-freedom dynamic model, and aerodynamic coefficient variable parameter model, it calls the feedback gain matrix and pole placement analysis results obtained from offline calculation to achieve compensation for multi-source disturbances and attitude stability control. The actuator drive unit is used to drive the actions of each actuator according to the attitude control commands output by the flight control processing unit, and, in conjunction with the output of the adaptive inverse compensation module, corrects the input distortion caused by electromagnetic interference or dynamic deviation of the actuator to ensure that the control commands are executed accurately.

[0014] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the above-described method.

[0015] Example 1 This embodiment is designed for drones flying in special environments such as wind turbulence, dust particle impact, and electromagnetic interference. The goal is to achieve stable attitude control and desired attitude tracking of the drone under conditions of multi-source disturbances. The system structure used in this embodiment is as follows: Figure 2As shown, it includes an attitude measurement unit, an environmental information acquisition unit, a flight control processing unit, and an actuator drive unit. The attitude measurement unit is used to obtain state information such as the attitude quaternion and angular velocity of the UAV. The environmental information acquisition unit is used to acquire or estimate environmental parameters such as wind speed, wind direction, dust particle concentration, and electromagnetic interference intensity. The flight control processing unit runs the attitude control algorithm of the present invention based on the above information. The actuator drive unit drives the servo motors, motors, and other actuators according to the control commands output by the flight control processing unit, which act on the UAV body and form a feedback closed loop through the attitude measurement unit. Supported by the system architecture, this embodiment first performs a unified modeling of special environmental disturbances; combined with Figure 3 The multi-source disturbance environment includes three main components: wind field disturbance, dust particle impact, and electromagnetic interference. In this embodiment, the additional aerodynamic forces and aerodynamic moments caused by the wind field are denoted as... , The impact of sand and dust particles is equivalent to the impact force and impact torque. , The effect of electromagnetic interference on the actuator channel is equivalent to the gain perturbation and additional noise of the control input, and its equivalent moment is denoted as... By superposition, the total disturbance force and disturbance moment can be obtained. , This allows multi-source disturbances in complex environments to be uniformly represented as time-varying disturbance vectors: This provides a clear disturbance path for subsequent dynamic modeling and robust control design; In terms of dynamic modeling, this embodiment uses quaternions. Describing the attitude of the UAV; in the inertial coordinate system, the six-degree-of-freedom attitude dynamics of the UAV under a disturbed environment can be expressed as: The attitude quaternion satisfies: The above equations constitute the basic model of nonlinear attitude dynamics in this embodiment; Considering that wind and dust can significantly alter the aerodynamic characteristics of the wings and control surfaces, this embodiment introduces an environmental parameter vector based on the traditional aerodynamic model. A variable structure model of aerodynamic coefficients is constructed to vary with flight state and environmental parameters; for example, the lift coefficient, pitch moment coefficient, and yaw moment coefficient can be expressed as: In terms of constructing control variables, this embodiment uses the desired attitude quaternion. and expected angular velocity For reference, we introduce attitude error quaternions: Its vector part is denoted as Based on this, construct the error state vector: In the flight control processing unit, a quadratic Lyapunov function is selected: Substituting the above error definition, six-degree-of-freedom dynamic model, and aerodynamic variable structure model into the equation, we can obtain the error system under multi-source disturbances. The upper bound of the derivative under the action is as follows: From this, we can derive the selection conditions for the control gain and the observer gain, so that the error state remains asymptotically stable or uniformly bounded under the premise that the disturbance is bounded. This provides stability constraints for the subsequent design of the control law and the disturbance observer. In terms of control law design, this embodiment implements a nonlinear attitude controller based on attitude error and angular velocity error in the flight control processing unit; the nominal control quantity is expressed by the following formula: To further suppress the equivalent external disturbances caused by sudden wind field changes, sandstorm impacts, and electromagnetic interference, this embodiment introduces a sliding mode disturbance observer outside the attitude controller. The sliding mode observer takes the actuator drive signal and the UAV's attitude, angular velocity, and other state information as inputs, and designs a disturbance estimation law: Based on disturbance estimation, feedforward compensation is introduced to obtain the final control command: That is, the nominal control quantity generated by the attitude controller is subtracted from the disturbance estimate given by the sliding mode observer to achieve active compensation for external disturbances and actuator uncertainties; In terms of closed-loop performance evaluation and parameter tuning, this embodiment also linearizes the nonlinear error system near the operating point to obtain a linear approximation model: And adopt status feedback Form a closed-loop matrix. To analyze the stability region of the control system under multi-source disturbances, this embodiment selects a parameter vector representing environmental and structural uncertainties: in This represents the average wind speed. Indicates turbulence intensity. The parameter grid is formed by representing the density of sand particles, the delay representing the overall delay of the actuator, and giving the expected range and step size of each parameter. For each set of values ​​in the parameter grid Calculate the corresponding closed-loop matrix. The eigenvalues ​​are then checked to see if their real parts satisfy the following: Simultaneously verify whether the sliding mode observer gain meets the disturbance suppression condition; if all the above conditions are met, mark the parameter point as a stable point and record the corresponding environmental parameters and control gain combination; by scanning all parameter points, the set of parameters that meet the stability requirements can be obtained, and its envelope region is regarded as the stability domain of the attitude control system under multi-source disturbances in this embodiment; in practical applications, the flight control processing unit preferentially selects the control gain located inside the stability domain and taking into account the dynamic performance index as the final tuning result.

[0016] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A method for attitude stabilization control of a UAV under special environmental disturbances, characterized in that, Includes the following steps: S1: Establish wind field disturbance model, dust particle impact model and electromagnetic interference model, and dynamically couple the three into the total disturbance vector experienced by the UAV; S2: Construct a six-degree-of-freedom dynamic model that includes the influence of the total disturbance vector. The model is used to describe the translational motion of the center of mass and the rotational motion around the center of mass of the UAV under multi-source disturbances in a complex environment. S3: Based on the traditional aerodynamic model, and by introducing environmental disturbance parameters, a variable structure model of aerodynamic coefficients is constructed. S4: Based on the model in steps S2 and S3, the Lyapunov function is constructed by using the error state vector and analyzing the convergence of attitude error, and its time derivative is derived to obtain the parameter constraints that make the system stable or uniformly ultimately bounded under multi-source bounded perturbations. S5: Under the constraints of step S4, the output of the attitude control law is combined with the disturbance estimate of the observer, and the multiplicative and additive electromagnetic interference in the actuator is corrected by the adaptive inverse compensation module to generate the actual command to drive the actuator. S6. Linearize the nonlinear closed-loop control system formed in step S5 near the desired equilibrium point to obtain a linear state-space model; by analyzing the distribution of the closed-loop system poles when the key environment and system parameters change, determine whether it meets the preset stability margin, thereby determining the stable operating region of the system; if the poles exceed the stable region, adjust the feedback gain matrix to reconfigure the poles.

2. The method according to claim 1, characterized in that: In step S1, based on the wind speed, particulate matter concentration and electromagnetic field intensity data collected in real time by the environmental information acquisition unit, the gust parameters of the wind field disturbance model, the particle distribution parameters of the sand and dust particle impact model and the noise parameters of the electromagnetic interference model are corrected online.

3. The method according to claim 1, characterized in that: In step S1, the total disturbance vector is constructed by integrating the total disturbance force and the total disturbance torque that change dynamically with time. The total disturbance force is composed of the vector sum of the aerodynamic force and the impact force, and the total disturbance torque is composed of the vector sum of the aerodynamic torque, the impact torque, and the equivalent torque.

4. The method according to claim 1, characterized in that: In step S2, the six-degree-of-freedom dynamic model integrates the UAV mass, center of mass velocity vector, gravitational acceleration vector, rotation matrix updated in real time by attitude quaternions, propulsion aerodynamic force, wind field additional force, and sand and dust particle impact force. The inertia matrix, body angular velocity vector, aerodynamic torque, wind torque, particle impact torque, and electromagnetic interference torque are all updated in real time with the evolution of the attitude quaternion, ensuring dynamic matching between dynamic characteristics and environmental disturbances.

5. The method according to claim 1, characterized in that: In step S3, the environmental disturbance parameters from step S1 are used as variables for aerodynamic coefficients, and the lift coefficient, pitch moment coefficient, and yaw moment coefficient are dynamically adjusted.

6. The method according to claim 1, characterized in that: In the model of step S3, the angle of attack gain, sideslip gain and environmental disturbance parameters are linked, the aerodynamic damping derivative is corrected in real time by the coupling change of the random body angular velocity and environmental disturbance, and the dynamic pressure, airspeed and wind field disturbance are matched in real time for the aerodynamic characteristics to adapt to special environments.

7. The method according to claim 1, characterized in that: In step S5, the attitude control law design adopts a composite structure of proportional term, differential term and environmental compensation term. The proportional term and differential term correspond to the real-time correction of attitude error and angular velocity error, respectively. The environmental compensation term directly calls the real-time output of the disturbance model in step S1 to offset the disturbance effect in advance. Observer design: A sliding mode observer architecture is adopted. By designing a nonlinear sliding surface and an adaptive sliding mode gain, finite-time convergence estimation of wind field abrupt changes, dust impacts, and electromagnetic interference is achieved. The adaptive inverse compensation module addresses command distortion caused by electromagnetic interference by constructing an inverse model of the actuator input based on the multiplicative disturbance estimate and additive noise estimate from the observer output. Through identity matrix normalization, it corrects the control command in real time.

8. A UAV attitude stabilization control system for special environmental disturbances, implementing the method described in any one of claims 1-7, characterized in that, include: The attitude measurement unit is used to collect inertial measurement information such as attitude quaternions and angular velocity of the UAV, providing input for attitude error calculation and dynamic state update; The environmental information acquisition unit is used to acquire or estimate wind field disturbance parameters, dust particle concentration, electromagnetic interference intensity, and other environmental disturbance information that affects aerodynamic changes. It provides a basis for disturbance modeling and parameter estimation by continuously collecting environmental data. The flight control processing unit is used to perform attitude error calculation, disturbance observer update, attitude control law solution and control command generation in real time based on the data provided by the attitude measurement unit and the environmental information acquisition unit during the flight of the UAV. The actuator drive unit is used to drive the actions of each actuator according to the attitude control commands output by the flight control processing unit, and, in conjunction with the output of the adaptive inverse compensation module, corrects the input distortion caused by electromagnetic interference or dynamic deviation of the actuator to ensure that the control commands are executed accurately.

9. The system according to claim 8, characterized in that: The flight control processing unit, based on a pre-built disturbance environment model, a six-degree-of-freedom dynamic model, and a variable parameter model of aerodynamic coefficients, calls the feedback gain matrix and pole placement analysis results obtained from offline calculations to achieve compensation for multi-source disturbances and attitude stabilization control.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method steps as described in any one of claims 1-7.

Images