MESO-based quadrotor unmanned aerial vehicle attitude control method and system
By decomposing and actively compensating for disturbance compensation errors using an improved Extended State Observer (MESO), the problem of insufficient disturbance compensation accuracy of traditional observers in the attitude control of quadcopter UAVs is solved, achieving high precision and stability improvement, and adapting to complex disturbance environments.
Patent Information
- Application Number
- CN202511721620.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-21
- Publication Date
- 2026-02-10
AI Technical Summary
In the existing technology, traditional extended state observers have insufficient perturbation compensation accuracy in the attitude control of quadcopter UAVs, significant peak phenomena, poor system adaptability, and difficulty in meeting the requirements of high precision and stability in complex perturbation environments.
An improved extended state observer (MESO) is adopted. By decomposing the disturbance compensation error into state estimation error and output estimation error, an output estimation error compensation channel is introduced, and active compensation is performed through compensation gain. Stability conditions are designed, and parameters are optimized to ensure the relationship between the observer bandwidth of the observer and the compensation channel gain, so as to achieve accurate handling of disturbances.
It improves the accuracy of disturbance compensation and tracking, enhances system robustness, adapts to complex disturbance scenarios, effectively suppresses peak phenomena, and improves system stability and adaptability.
Smart Images
Figure CN121501009A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of automatic control technology, and in particular to a method and system for attitude control of a quadcopter UAV based on MESO, which is applicable to nonlinear systems with external disturbances and unmodeled dynamics, such as UAVs, robots, permanent magnet synchronous motors and other high-performance servo systems. Background Technology
[0002] In fields such as robotics, aerospace, and precision manufacturing that rely on high-precision servo control, nonlinear systems generally face the dual impact of external disturbances and internal unmodeled dynamics. External disturbances include wind disturbances from quadcopter drones and load fluctuations from robots, while internal unmodeled dynamics include system friction and parameter drift. These disturbances directly lead to a decrease in system tracking accuracy and a weakening of stability, becoming the core bottleneck restricting high-performance control.
[0003] Among existing solutions, the proposed Active Disturbance Rejection Control (ADRC) and its core component, the Extended State Observer (ESO), offer a solution by treating unmodeled internal dynamics and external disturbances as a unified total disturbance and compensating for them in real time. Traditional ESOs suffer from the following drawbacks: 1. They rely solely on a single observation channel to estimate the total disturbance, failing to reveal the mechanism underlying the disturbance compensation error, which is actually determined by both state estimation and output estimation errors. They cannot specifically eliminate error sources, making it difficult to meet the high-precision requirements of scenarios such as quadcopter UAV attitude control and precision platform positioning. 2. When the initial state of the observer deviates significantly from the actual system state, high-gain design can lead to substantial overshoot in state estimation during the transition process, potentially causing system oscillations or even compromising system stability. 3. Subsequent improvements, such as nonlinear function-optimized ESOs, cascaded ESOs, and finite-time convergent ESOs, while attempting to optimize the observer structure, focus on the observer's own dynamic adjustments, neglecting solutions for disturbance compensation error decomposition and proactive compensation of output estimation errors. These solutions cannot simultaneously address accuracy and peak performance issues, making them unsuitable for high-performance servo systems operating under complex disturbance environments. Summary of the Invention
[0004] The main objective of this invention is to provide a method and system for attitude control of a quadcopter UAV based on MESO, which solves the technical problems of insufficient disturbance compensation accuracy, significant peak phenomenon, and poor system adaptability in the prior art.
[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a quadcopter UAV attitude control method based on an improved extended state observer, comprising the following steps: Mathematical modeling of nonlinear systems yields state equations for control systems with disturbances. An improved extended state observer is designed based on the state equation of the control system. The improved extended state observer includes a state observer and an output estimation error compensation channel. The total system disturbance is estimated by an improved extended state observer and compensated for in the control law to suppress periodic disturbances.
[0006] In a preferred embodiment, the design of the improved extended state observer includes: Establish a decomposition structure for the disturbance compensation error, decomposing the disturbance compensation error into state estimation error and output estimation error; An output estimation error compensation channel is introduced to actively compensate for the output estimation error through compensation gain; The compensation gain satisfies a stability condition, wherein the stability condition is determined based on the relationship between the state observer gain and the compensation channel gain. The improved extended state observer achieves precise processing of disturbance compensation errors through the decomposition structure and the compensation channel; The parameters are adjusted based on the state observer gain and the compensation channel gain.
[0007] In the preferred embodiment, the compensation gain k of the output estimation error compensation channel satisfies the stability condition, expressed as: ; in, For state observer gain; Compensation channel gain.
[0008] The preferred option also includes: The output estimation error is adjusted in real time based on the compensation gain to generate a compensation signal; The compensation signal is fed back to the improved extended state observer to optimize the total disturbance estimation; If the output estimation error exceeds a preset threshold, the compensation effect is amplified by the compensation gain. The difference between the state estimation error and the output estimation error is obtained to further refine the compensation process.
[0009] In the preferred embodiment, estimating the total system disturbance using the improved extended state observer and compensating for it in the control law includes: The improved extended state observer is applied to a nonlinear system, wherein the periodic disturbance includes external disturbances such as wind disturbance; the nonlinear system is a quadcopter unmanned aerial vehicle attitude control system. The state variables and output variables of the nonlinear system are obtained, and preliminary estimation is performed using the state observer. The estimation results are adjusted according to the output estimation error compensation channel to obtain an accurate total disturbance estimate. The nominal control input and the total disturbance estimate are introduced into the control law, and compensation calculation is performed using the system input gain. The control law is designed as the nominal control input minus the total disturbance estimate divided by the system input gain, with an output estimation error term added.
[0010] In the preferred embodiment, the step of mathematically modeling the nonlinear system to obtain the state equations of the control system with disturbances includes: A state-space model is established for a nonlinear system, where the state variables include the system output and its derivative, and the total disturbance includes the nonlinear function and the unknown disturbance. The nonlinear term and the unknown disturbance are treated as a total disturbance, and a differential term of the total disturbance is introduced; The state equations of the control system are constructed based on the system input gain and control input. If the nonlinear system is a quadcopter UAV attitude control system, then the state equation incorporates attitude angle and angular velocity variables; The disturbance influence parameters are obtained through the mathematical modeling.
[0011] In the preferred embodiment, the nominal control input and the estimated total disturbance are introduced into the control law, and compensation is calculated using the system input gain, as shown in the formula: ; in, For nominal control input, This is the estimated total disturbance. To output the estimation error, Input gain to the system; Obtain the nominal control input, which is determined based on the desired system response; Extract the total disturbance estimate and output the estimation error from the improved extended state observer; The total disturbance estimate is normalized by the system input gain to generate a compensation term; the compensation term is then combined with the nominal control input to form the final control law. If there is an external disturbance, the compensation strength is dynamically adjusted based on the output estimation error.
[0012] In the preferred embodiment, the step of establishing a decomposition structure for the disturbance compensation error, which decomposes the disturbance compensation error into state estimation error and output estimation error, includes: The components of the disturbance compensation error are analyzed and decomposed into state estimation error component and output estimation error component. The state estimation error represents the observer's deviation from the system state, and the output estimation error represents the difference between the observed output and the actual output. The decomposition structure reveals the sources of error and provides a basis for compensation strategies. An independent adjustment mechanism is introduced to address the state estimation error. If the output estimation error dominates the total error, it should be processed first through the compensation channel; The parameter design of the improved extended state observer is optimized based on the decomposition results.
[0013] In a preferred embodiment, the improved extended state observer includes a state observer and an output estimation error compensation channel, comprising: The state observer is designed with a gain based on a bandwidth configuration method, satisfying a specific relationship between the gain and the observer bandwidth. The expression is: ; in, The observer bandwidth represents the frequency response parameter of the observer. Prove the global uniform eventual bounded stability of closed-loop systems using Lyapunov theory; The output estimation error compensation channel independently adjusts the disturbance compensation process; The performance of the improved extended state observer was compared with that of the traditional extended state observer in the experimental verification.
[0014] Secondly, the present invention provides an attitude control system for a quadcopter unmanned aerial vehicle based on MESO, comprising the following steps: The mathematical modeling module is used to perform mathematical modeling on nonlinear systems and obtain the state equations of control systems with disturbances. The MESO module is used to design an improved extended state observer based on the state equation of the control system. The improved extended state observer includes a state observer and an output estimation error compensation channel. The disturbance compensation module is used to estimate the total system disturbance through the improved extended state observer and to compensate for it in the control law to suppress periodic disturbances.
[0015] This invention provides an attitude control method for a quadrotor UAV based on an improved extended state observer (EMS). The method includes: mathematically modeling the nonlinear system to obtain the state equations of the control system with disturbances; designing a Multi-Effect State Observer (MESO) based on the state equations to decompose the disturbance compensation error into two parts: state estimation error and output estimation error, thus improving the accuracy of disturbance estimation; estimating the total system disturbance using MESO and compensating it in the control law; establishing a disturbance compensation error decomposition framework for the system; realizing attitude control; and providing real-time compensation for periodic disturbance compensation errors, effectively suppressing peak phenomena. This improves disturbance compensation accuracy and tracking accuracy, enhances system robustness, and adapts to complex disturbance scenarios. Attached Figure Description
[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the control method of the present invention; Figure 2 This is a control block diagram of the MESO of the present invention; Figure 3 This is a comparison chart of the disturbance suppression effects of traditional ESO and MESO in this invention. Detailed Implementation
[0017] Example 1 like Figure 1-3 As shown, a method for attitude control of a quadcopter UAV based on a Modified Extended State Observer (MESO) includes the following steps: S1: Mathematically model the nonlinear system to obtain the state equations of the control system with disturbances.
[0018] S2: Design MESO based on the state equation of the control system. MESO includes a state observer and an output estimation error compensation channel.
[0019] S3: Estimate the total system disturbance using MESO and compensate for it in the control law to suppress periodic disturbances.
[0020] This embodiment uses nonlinear system mathematical modeling to obtain the state equation of the control system with disturbances. Then, MESO is designed for the state equation to decompose the disturbance compensation error into two parts: state estimation error and output estimation error, which improves the accuracy of disturbance estimation. The total system disturbance is estimated by MESO and compensated in the control law, thus establishing the system's disturbance compensation error decomposition framework, realizing attitude control, and real-time compensation for periodic disturbance compensation error, effectively suppressing peak phenomena. This improves the disturbance compensation accuracy and tracking accuracy, enhances the disturbance estimation accuracy and system robustness, and adapts to complex disturbance scenarios.
[0021] This embodiment introduces a measurable state estimation error compensation mechanism to suppress peak phenomena and enhance the robustness and stability of the system. The following steps will describe the process.
[0022] S1: Mathematical modeling of nonlinear systems. Taking the attitude control of the roll channel of a quadcopter UAV as an example, a second-order nonlinear system model is established, a decomposition framework is established, and it is revealed that the disturbance compensation error consists of two components: disturbance estimation error and state estimation error, providing a theoretical basis for the compensation strategy.
[0023] In the preferred embodiment, a mathematical model is performed on the nonlinear system to obtain the state equations of the control system with disturbances, including: A state-space model is established for the nonlinear system, where the state variables include the system output and its derivative, and the total disturbance includes the nonlinear function and the unknown disturbance.
[0024] The nonlinear term and the unknown disturbance are treated as a total disturbance, and a differential term of the total disturbance is introduced.
[0025] Construct the state equations of the control system based on the system input gain and control input.
[0026] If the nonlinear system is a quadcopter UAV attitude control system, then the state equation incorporates attitude angle and angular velocity variables.
[0027] The parameters of the disturbance effect are obtained through mathematical modeling and used for subsequent observer design.
[0028] This embodiment decomposes the disturbance compensation error into two components: state estimation error and output estimation error, providing a theoretical basis for accurate compensation and improving the accuracy of disturbance estimation. It also verifies the integrity of the control system's state equations, ensuring coverage of periodic disturbance characteristics.
[0029] According to the UAV dynamics equations, the attitude angle of the roll channel (Corresponding system status) ) and attitude angular rate (Corresponding system status) The dynamic relationship of ) satisfies: ,in, The length of the drone's arm. The moment of inertia of the rolling channel. , These are the lift torques of the front and rear motors, respectively. The damping nonlinearity of the roll channel (corresponding to the system nonlinear function) ), Roll channel disturbance caused by wind disturbance (corresponding to unknown disturbance) ).
[0030] In this embodiment, the state space of the nonlinear system is considered as follows: a second-order nonlinear system, with the following formula: ; in, For the system's state, To control input, For system output, For nonlinear functions, For unknown disturbances and This is the system's input gain.
[0031] Within the ADRC framework, nonlinear terms and unknown disturbances Considered as a "total disturbance", denoted as: .
[0032] In the preferred scheme, the system state space reconstruction can be written in the following state space form, with the expression: ; in, This represents the differential of the total disturbance.
[0033] In the preferred scheme, a decomposition structure for the disturbance compensation error is established, which decomposes the disturbance compensation error into state estimation error and output estimation error, including: The components of the disturbance compensation error are analyzed and decomposed into state estimation error component and output estimation error component. The state estimation error represents the observer's deviation from the system state, and the output estimation error represents the difference between the observed output and the actual output. The decomposition structure reveals the sources of error and provides a basis for compensation strategies.
[0034] An independent adjustment mechanism is introduced to address the state estimation error.
[0035] If the output estimation error dominates the total error, then it should be processed first through the compensation channel.
[0036] The parameter design of MESO was optimized based on the decomposition results.
[0037] This embodiment improves the overall accuracy of disturbance estimation by decomposition, ensuring the effectiveness of compensation.
[0038] Step S2: MESO parameter design 1. State observer gain design: based on the response requirements of the quadcopter UAV's roll channel. In the preferred embodiment, MESO includes a state observer and an output estimation error compensation channel, including: The state observer's gain is designed based on a bandwidth configuration method to satisfy a specific relationship between the gain and the observer's bandwidth. The expression is: ; in, The observer bandwidth is denoted by , which represents the observer's frequency response parameter.
[0039] The output estimation error compensation channel independently adjusts the disturbance compensation process.
[0040] The performance of MESO and traditional ESO was compared in the experimental verification.
[0041] Based on the results, MESO was found to have advantages in terms of perturbation estimation accuracy, tracking precision, and anti-interference capability.
[0042] Extend MESO to other nonlinear systems, such as robots or aerospace vehicles.
[0043] This embodiment designs the state observer gain based on a bandwidth configuration method, clarifies the physical meaning of the parameters, simplifies the engineering implementation process, and improves the intuitiveness of the parameter design criteria.
[0044] 2. Design of compensation gain k In this embodiment, when In this case, the degree of freedom is increased in the disturbance compensation error, which can reduce the state estimation error.
[0045] The preferred option also includes: The output estimation error is adjusted in real time based on the compensation gain to generate a compensation signal.
[0046] The compensation signal is fed back into MESO to optimize the total disturbance estimate.
[0047] If the output estimation error exceeds the preset threshold, the compensation effect is amplified by the compensation gain.
[0048] The difference between the state estimation error and the output estimation error is obtained and used to further refine the compensation process.
[0049] Active compensation mechanisms are used to reduce peak phenomena and ensure a smooth transition in the estimation process.
[0050] In this embodiment, when At that time, the state estimation error was fully compensated, and the peak phenomenon was eliminated.
[0051] In the preferred scheme, the compensation gain k of the output estimation error compensation channel satisfies the stability condition, expressed as: ; in, For state observer gain; Compensation channel gain.
[0052] This embodiment actively compensates for the output estimation error by compensating for the gain k, which satisfies the stability condition and achieves effective compensation, reducing the state estimation error. When the stability condition of k>l2 / b0 is met, the state estimation error is fully compensated, effectively suppressing the peak phenomenon and improving the transient performance of the system.
[0053] 3. Linear MESO construction: Based on the above parameters, a linear MESO is constructed to estimate the roll channel state and total disturbance.
[0054] In the preferred embodiment, the design of MESO includes: A decomposition structure for the disturbance compensation error is established, which decomposes the disturbance compensation error into state estimation error and output estimation error.
[0055] An output estimation error compensation channel is introduced to actively compensate for the output estimation error through compensation gain.
[0056] The compensation gain satisfies the stability condition, which is determined based on the relationship between the state observer gain and the compensation channel gain.
[0057] MESO achieves precise processing of disturbance compensation errors by decomposing the structure and compensation channels.
[0058] The parameters are adjusted based on the state observer gain and the compensation channel gain to ensure the stability and accuracy of the observation process.
[0059] The disturbance in the state equation of the control system is estimated in real time by MESO, and the total disturbance estimate is used for subsequent compensation.
[0060] In this embodiment, the linear MESO design formula used to estimate the system state is: ; In the formula, for Discretized estimate, This formula is used to discretize the output measurements and to calculate the estimated values of the system state and total disturbance in real time.
[0061] like Figure 2 The diagram shown illustrates the structure and signal interaction of MESO, illustrating its internal components and their relationship with the controlled object and control law. The core of MESO comprises three main modules and two types of key signal flows: 1. Core Module Controlled object: Nonlinear system represented by the attitude control system of quadcopter UAV, the input is the control signal, the output is the actual state of the system (such as attitude angle), and it is also affected by external disturbances.
[0062] MESO main body: Composed of a state observer and an output estimation error compensation channel. The state observer receives the output y(t) of the controlled object and combines it with the control signal. The system state is estimated by preset gains l1, l2, and l3. , and total disturbance The output estimation error compensation channel calculates the actual output y(t) and the state estimate. The difference (output estimation error Δx1(t)) is used to generate a compensation signal through the compensation gain k, which is then fed back to the state observer. Estimate the channel; Control law module: Receives the total disturbance estimate from the MESO output. The final control signal is calculated by combining the output estimation error Δx1(t) with the nominal control input. Input the controlled object; 2. Signal flow Forward flow: Control law output →Controlled object→Output →MESO; Feedback Flow: MESO Output The control law Δx1(t) forms a closed-loop control, which improves the real-time performance and accuracy of disturbance compensation.
[0063] In this embodiment, a compensation gain is introduced to achieve independent adjustment of the state and disturbance compensation channels. The MESO parameters can be directly determined based on the system parameters, and when the derivation conditions are met, state estimation errors can be eliminated and peak phenomena can be effectively suppressed.
[0064] Step S3: Control Law Implementation and Disturbance Compensation In the preferred scheme, the total system disturbance is estimated using MESO and compensated for in the control law, including: MESO is applied to a nonlinear system, where periodic disturbances include external disturbances such as wind disturbances; the nonlinear system is the attitude control system of a quadcopter UAV. Obtain the state and output variables of the nonlinear system and perform preliminary estimation using a state observer; The estimation results are adjusted based on the output estimation error compensation channel to obtain an accurate total disturbance estimate. Introduce the nominal control input and the total disturbance estimate into the control law, and perform compensation calculations using the system input gain; The control law is designed as the nominal control input minus the total disturbance estimate divided by the system input gain, with an output estimation error term added.
[0065] The specific implementation is as follows: 1. Introduce the nominal control input and the estimated total disturbance into the control law, and calculate the compensation using the system input gain. The formula is as follows: ; in, For nominal control input, This is the estimated total disturbance. To output the estimation error, Input gain to the system; To increase the compensation channel gain.
[0066] Obtain the nominal control input, and generate the nominal control input using a PID controller based on the desired attitude angle of the roll channel; wherein the nominal control input is determined based on the desired system response; 2. Calculation of disturbance compensation control law Extract the total disturbance estimate and output the estimation error from MESO; The total disturbance estimate is normalized by the system input gain to generate a compensation term; the compensation term is then combined with the nominal control input to form the final control law. If there is an external disturbance, the compensation strength is dynamically adjusted based on the output estimation error.
[0067] 3. Control signal output The calculated u(t) is converted into a motor control signal and applied to the nonlinear system through a control law to achieve real-time suppression of periodic disturbances and stable system control. That is, the output is sent to the motor drive module of the quadcopter UAV to realize attitude control and disturbance compensation of the roll channel.
[0068] This embodiment employs a control law with a clear compensation mechanism. By introducing an output estimation error compensation term, it achieves real-time compensation for periodic interference compensation errors, thus achieving the expected suppression effect.
[0069] Step S4: System stability verification and performance testing. The closed-loop system is globally consistent and eventually bounded and stable, as proved by Lyapunov theory.
[0070] Hardware-in-the-loop experiments conducted on a quadcopter UAV platform show that, under a complex disturbance environment consisting of wind-like disturbances and step signals, by injecting a complex disturbance signal and comparing the performance of MESO and traditional ESO, MESO achieves higher disturbance compensation accuracy in roll, pitch, and yaw channels compared to traditional ESO, and effectively reduces peak overshoot.
[0071] like Figure 3 As shown, the performance advantages of MESO are intuitively demonstrated through the difference between the two curves: Traditional ESO curve: The overall fluctuation range is large, with obvious peak overshoot in the initial stage of the experiment (t=0-0.5s), with a maximum error of 3.2°, and the mean error is maintained at around 0.42° throughout the entire experimental period of about 12s. The response to the suppression of compound disturbances is lagging.
[0072] MESO curve: There is no obvious peak overshoot throughout the entire process. The error converges rapidly to below 0.5° in the initial stage. The error fluctuation is smooth throughout the entire experimental period, with an average value as low as 0.25°. It tracks and cancels periodic disturbances more promptly, and the error is always lower than that of the traditional ESO curve.
[0073] This indicates that under the same complex disturbance environment, MESO is significantly better than traditional ESO in terms of peak suppression, error amplitude, and response speed, verifying the superiority of MESO in disturbance suppression accuracy and system dynamic stability.
[0074] This embodiment is applicable to motion control systems such as quadcopter UAVs. It adopts the same parameter design logic and control law structure, and can achieve high-precision control of all attitude channels. Under periodic disturbance environments such as wind disturbance, it achieves higher disturbance compensation accuracy and tracking accuracy for each channel.
[0075] Example 2 To further illustrate with reference to Example 1, a quadcopter unmanned aerial vehicle (UAV) attitude control system based on MESO includes the following steps: The mathematical modeling module is used to perform mathematical modeling on nonlinear systems and obtain the state equations of control systems with disturbances. The MESO module is used to design a MESO based on the state equations of the control system. The MESO includes a state observer and an output estimation error compensation channel.
[0076] The disturbance compensation module is used to estimate the total system disturbance through MESO and compensate for it in the control law to suppress periodic disturbances.
[0077] This embodiment provides a working process, working details and technical effects of a MESO-based attitude control method for quadrotor UAVs. Please refer to Embodiment 1 for details, which will not be repeated here.
[0078] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.
Claims
1. A method for attitude control of a quadrotor UAV based on MESO, characterized in that, Includes the following steps: Mathematical modeling of nonlinear systems yields state equations for control systems with disturbances. An improved extended state observer is designed based on the state equation of the control system. The improved extended state observer includes a state observer and an output estimation error compensation channel. The total system disturbance is estimated by an improved extended state observer and compensated for in the control law to suppress periodic disturbances.
2. The attitude control method for a quadrotor UAV based on MESO according to claim 1, characterized in that, The design of the improved extended state observer includes: Establish a decomposition structure for the disturbance compensation error, decomposing the disturbance compensation error into state estimation error and output estimation error; An output estimation error compensation channel is introduced to actively compensate for the output estimation error through compensation gain; The compensation gain satisfies a stability condition, wherein the stability condition is determined based on the relationship between the state observer gain and the compensation channel gain. The improved extended state observer achieves precise processing of disturbance compensation errors through the decomposition structure and the compensation channel; The parameters are adjusted based on the state observer gain and the compensation channel gain.
3. The attitude control method for a quadrotor UAV based on MESO according to claim 2, characterized in that, The compensation gain k of the output estimation error compensation channel satisfies the stability condition, expressed as: ； in, For state observer gain; Compensation channel gain.
4. The attitude control method for a quadrotor UAV based on MESO according to claim 3, characterized in that, Also includes: The output estimation error is adjusted in real time based on the compensation gain to generate a compensation signal; The compensation signal is fed back to the improved extended state observer to optimize the total disturbance estimation; If the output estimation error exceeds a preset threshold, the compensation effect is amplified by the compensation gain. The difference between the state estimation error and the output estimation error is obtained to further refine the compensation process.
5. The attitude control method for a quadrotor UAV based on MESO according to claim 1, characterized in that, The step of estimating the total system disturbance using the improved extended state observer and compensating for it in the control law includes: The improved extended state observer is applied to a nonlinear system, wherein the periodic disturbance includes external disturbances such as wind disturbance; the nonlinear system is a quadcopter unmanned aerial vehicle attitude control system. The state variables and output variables of the nonlinear system are obtained, and preliminary estimation is performed using the state observer. The estimation results are adjusted according to the output estimation error compensation channel to obtain an accurate total disturbance estimate. The nominal control input and the total disturbance estimate are introduced into the control law, and compensation calculation is performed using the system input gain. The control law is designed as the nominal control input minus the total disturbance estimate divided by the system input gain, with an output estimation error term added.
6. The attitude control method for a quadrotor UAV based on MESO according to claim 1, characterized in that, The mathematical modeling of the nonlinear system, yielding the state equations of the control system with disturbances, includes: A state-space model is established for a nonlinear system, where the state variables include the system output and its derivative, and the total disturbance includes the nonlinear function and the unknown disturbance. The nonlinear term and the unknown disturbance are treated as a total disturbance, and a differential term of the total disturbance is introduced; The state equations of the control system are constructed based on the system input gain and control input. If the nonlinear system is a quadcopter UAV attitude control system, then the state equation incorporates attitude angle and angular velocity variables; The disturbance influence parameters are obtained through the mathematical modeling.
7. The attitude control method for a quadrotor UAV based on MESO according to claim 1, characterized in that, The nominal control input and the estimated total disturbance are introduced into the control law, and compensation is calculated using the system input gain, as shown in the formula: ； in, For nominal control input, This is the estimated total disturbance. To output the estimation error, Input gain to the system; Obtain the nominal control input, which is determined based on the desired system response; Extract the total disturbance estimate and output the estimation error from the improved extended state observer; The total disturbance estimate is normalized by the system input gain to generate a compensation term; the compensation term is then combined with the nominal control input to form the final control law. If there is an external disturbance, the compensation strength is dynamically adjusted based on the output estimation error.
8. The attitude control method for a quadrotor UAV based on MESO according to claim 2, characterized in that, The decomposition structure for establishing the disturbance compensation error decomposes the disturbance compensation error into state estimation error and output estimation error, including: The components of the disturbance compensation error are analyzed and decomposed into state estimation error component and output estimation error component. The state estimation error represents the observer's deviation from the system state, and the output estimation error represents the difference between the observed output and the actual output. The decomposition structure reveals the sources of error and provides a basis for compensation strategies. An independent adjustment mechanism is introduced to address the state estimation error. If the output estimation error dominates the total error, it should be processed first through the compensation channel; The parameter design of the improved extended state observer is optimized based on the decomposition results.
9. The attitude control method for a quadrotor UAV based on MESO according to claim 1, characterized in that, The improved extended state observer includes a state observer and an output estimation error compensation channel, including: The state observer is designed with a gain based on a bandwidth configuration method, satisfying a specific relationship between the gain and the observer bandwidth. The expression is: ； in, The observer bandwidth represents the frequency response parameter of the observer. Prove the global uniform eventual bounded stability of closed-loop systems using Lyapunov theory; The output estimation error compensation channel independently adjusts the disturbance compensation process; The performance of the improved extended state observer was compared with that of the traditional extended state observer in the experimental verification.
10. An attitude control system for a quadcopter unmanned aerial vehicle based on MESO, characterized in that, Includes the following steps: The mathematical modeling module is used to perform mathematical modeling on nonlinear systems and obtain the state equations of control systems with disturbances. The MESO module is used to design an improved extended state observer based on the state equation of the control system. The improved extended state observer includes a state observer and an output estimation error compensation channel. The disturbance compensation module is used to estimate the total system disturbance through the improved extended state observer and to compensate for it in the control law to suppress periodic disturbances.