Unmanned aerial vehicle wind resistance control method based on improved LQR-LADRC

Abstract

The invention relates to an unmanned aerial vehicle wind resistance control method based on improved LQR-LADRC. The method comprises the following steps: constructing an unmanned aerial vehicle attitude dynamic model containing disturbance terms; based on the unmanned aerial vehicle attitude dynamic model, constructing an attitude controller based on improved LQR-LADRC; based on the attitude controller, calculating a system control quantity capable of driving the attitude of the unmanned aerial vehicle to be adjusted to an expected state; and performing unmanned aerial vehicle wind resistance control based on the system control quantity. The method can improve the response speed of the attitude of the unmanned aerial vehicle, improves the interference suppression capability of the unmanned aerial vehicle, and enables the unmanned aerial vehicle to maintain stable in a complex wind interference environment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of unmanned aerial vehicles, and particularly to a wind resistance control method for unmanned aerial vehicles based on improved LQR-LADRC. Background Art

[0002] Under the development trends of large-scale, large-scale expansion, and deep-sea development of offshore wind turbines, problems such as high operation and maintenance costs and poor accessibility are becoming increasingly prominent, and traditional manual inspection methods are difficult to meet the requirements. With the development of multi-rotor unmanned aerial vehicle technology, multi-rotor unmanned aerial vehicles have been widely used in the field of wind power inspection. At present, relevant researchers have carried out research on the inspection scheme for offshore wind turbines without shutdown.

[0003] Wind farms are rich in wind energy resources and the wind conditions are complex and changeable. When an unmanned aerial vehicle performs an inspection task in a state of non-stop operation of a wind turbine, it will be affected by complex wind disturbances, resulting in a decrease in its stability. The current unmanned aerial vehicle control algorithms still have certain limitations in the application of complex wind disturbance environments. For example, PID control has the advantages of simple structure and independence from the system model, but for a system of a quadrotor unmanned aerial vehicle with nonlinearity, strong coupling, underactuation, and external wind disturbances, its control effect is not ideal.

[0004] Therefore, how to improve the interference suppression ability of unmanned aerial vehicles and make them maintain stability in a complex wind disturbance environment is still a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The purpose of the present invention is to provide a wind resistance control method for unmanned aerial vehicles based on improved LQR-LADRC, which can improve the response speed of the attitude of unmanned aerial vehicles and maintain stability in a complex wind disturbance environment through the interference suppression ability of unmanned aerial vehicles.

[0006] To achieve the above purpose, the present invention provides the following solution:

[0007] A wind resistance control method for unmanned aerial vehicles based on improved LQR-LADRC includes:

[0008] Constructing an unmanned aerial vehicle attitude dynamics model containing a disturbance term;

[0009] Based on the unmanned aerial vehicle attitude dynamics model, constructing an attitude controller based on improved LQR-LADRC; wherein, the attitude controller of improved LQR-LADRC includes: calculating the desired angular velocity based on the attitude angle error, designing an optimal control law based on the attitude angle error and the angular velocity error, and introducing a second-order extended state observer;

[0010] Based on the attitude controller of improved LQR-LADRC, calculating a system control quantity that can drive the attitude of the unmanned aerial vehicle to be adjusted to the desired state;

[0011] Perform anti-wind control of the UAV based on the system control quantity.

[0012] Optionally, the UAV attitude dynamics model is:

[0013]

[0014] Wherein, θ and ψ are the roll angle, pitch angle, and yaw angle of the UAV respectively, and I x , I y , I z are the moments of inertia about the x, y, and z axes of the airframe respectively. represents the roll angle angular acceleration. represents the pitch angle angular acceleration. represents the yaw angle angular acceleration. f θ , f ψ represent the total disturbances including internal system disturbances and external disturbances on the roll channel, pitch channel, and yaw channel respectively.

[0015] Optionally, based on the attitude controller of the improved LQR-LADRC, the system control quantities that can drive the UAV attitude to adjust to the desired state include:

[0016] Obtain the desired angular velocity by using P control for the attitude angle error, introduce the angular velocity error, design the optimal control law with the attitude angle error and angular velocity error as state variables, so that the attitude angle error and angular velocity error quickly converge to 0.

[0017] Introduce a second-order extended state observer to estimate the total disturbance and perform feedforward compensation, and combine with the optimal control law to obtain the system control quantity.

[0018] Optionally, the optimal control law includes: the optimal control law for the roll angle channel, the optimal control law for the pitch angle channel, and the optimal control law for the yaw angle channel.

[0019] Optionally, designing the optimal control law includes:

[0020] Design the optimal control law for the roll angle channel, including:

[0021] Use P control for the error between the desired roll angle and the actual roll angle to obtain the desired roll angular velocity; the desired roll angular velocity is:

[0022]

[0023] Wherein, is the desired attitude angle, output by the UAV position controller. is the proportional coefficient of P control;

[0024] Based on the output of the position controller is a constant value, then the expected roll angle change rate is taken as 0, that is:

[0025]

[0026] Take the roll angle error and roll angular velocity error as state variables and

[0027]

[0028] in, is the angular velocity of the roll angle;

[0029] Take control quantity for:

[0030]

[0031] Based on the expected roll rate Expected roll angle change rate is 0, roll angle error and roll angle velocity error, control amount The state space equation for obtaining the roll angle channel is:

[0032]

[0033] That is, the roll angle channel system matrix is:

[0034]

[0035] The control matrix is:

[0036]

[0037] The state variable weight matrix is taken as:

[0038]

[0039] The control variable weight matrix is taken as:

[0040]

[0041] By solving the Riccati equation, we can calculate the matrix

[0042]

[0043] Matrix-based Calculate the roll channel feedback gain matrix

[0044]

[0045] The optimal control rate of the roll angle channel is obtained as:

[0046]

[0047] in, is the optimal control rate of the roll angle channel, and is the gain matrix element.

[0048] Optionally, designing the optimal control rate of the pitch angle channel includes:

[0049] The controller parameters of the pitch angle channel are designed in common with the roll angle channel; the optimal control rate of the pitch angle channel is:

[0050]

[0051] Among them, u 0θ is the optimal control rate of the pitch angle channel.

[0052] Optionally, designing the optimal control rate of the yaw angle channel includes:

[0053] The design process of the optimal control rate of the yaw angle channel is consistent with the design process of the optimal control rate of the roll angle channel; the optimal control rate of the yaw angle channel is:

[0054]

[0055] Among them, u 0ψ is the optimal control rate of the pitch angle channel, is the yaw channel feedback gain matrix, x 1ψ =ψ d -ψ is the yaw angle error state variable, is the yaw angle angular velocity error state variable, is the yaw angular velocity.

[0056] Optionally, the second-order extended state observer includes: a second-order extended state observer for roll angle, a second-order extended state observer for pitch angle, and a second-order extended state observer for yaw angle;

[0057] The second-order extended state observer of the roll angle is:

[0058]

[0059] in, Roll angle Roll angle angular velocity Total disturbance in roll channel The estimated value of are their rates of change, is the roll angle and pitch angle observer bandwidth;

[0060] The second-order extended state observer of the pitch angle is:

[0061]

[0062] Among them, z 1θ 、z 2θ 、z 3θ They are the pitch angle θ and the pitch angle angular velocity respectively Total disturbance of pitch channel f θ The estimated value of are their rates of change, is the roll angle and pitch angle observer bandwidth;

[0063] The second-order extended state observer of the yaw angle is:

[0064]

[0065] Among them, z 1ψ 、z 2ψ 、z 3ψ are the yaw angle ψ and the yaw angular velocity respectively Total disturbance in yaw channel f ψ The estimated value of are their rates of change, w oψ is the yaw angle observer bandwidth.

[0066] Optionally, the system control variable is:

[0067]

[0068] in, Indicates the control quantity of the roll channel system, u θ Indicates the control quantity of the pitch channel system, u ψ represents the control quantity of the yaw channel system, represents the roll channel disturbance compensation factor, b θ represents the pitch channel disturbance compensation factor, b ψ Represents the yaw channel disturbance compensation factor.

[0069] The beneficial effects of the present invention are:

[0070] The improved LQR of the present invention can respond to the roll angle error more quickly, thereby reducing the steady-state error of the system under disturbance; the introduction of LESO can improve the interference suppression capability of the system and further reduce the steady-state error; compared with the existing controller, the maximum steady-state error value of the improved LQR-LADRC controller of the present invention is smaller and has better interference suppression effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0072] Figure 1 Schematic diagram of a flow chart of a UAV wind resistance control method based on improved LQR-LADRC according to an embodiment of the present invention;

[0073] Figure 2 Schematic diagram of the structure of the improved LQR-LADRC controller according to an embodiment of the present invention;

[0074] Figure 3 Response diagrams of four controllers, namely, improved LQR-LADRC, LQR-LADRC, improved LQR, and LQR, under no disturbance according to an embodiment of the present invention;

[0075] Figure 4 Response diagrams of four controllers, namely, improved LQR-LADRC, LQR-LADRC, improved LQR, and LQR, under sinusoidal disturbance according to an embodiment of the present invention. DETAILED DESCRIPTION

[0076] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0077] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0078] like Figure 1 As shown, this embodiment proposes a UAV wind resistance control method based on improved LQR-LADRC, including:

[0079] Construct a UAV attitude dynamics model with disturbance terms;

[0080] Based on the attitude dynamics model of the UAV, an attitude controller based on improved LQR-LADRC is constructed; among them, the attitude controller of improved LQR-LADRC includes: calculating the desired angular velocity based on the attitude angle error, designing the optimal control law based on the attitude angle error and the angular velocity error, and introducing a second-order extended state observer; as Figure 2 shown;

[0081] Based on the attitude controller of the improved LQR-LADRC, the system control quantity capable of driving the UAV attitude to adjust to the desired state is calculated;

[0082] Based on the system control quantity, wind resistance control of the UAV is carried out.

[0083] Specifically, in this embodiment, constructing the UAV attitude dynamics model including disturbance terms includes:

[0084] First, construct the UAV attitude dynamics model:

[0085]

[0086] Classify the internal disturbances such as strong coupling terms and external disturbances as the total disturbance f θ 、f ψ , expressed as:

[0087]

[0088] Take the control quantity u θ ,u ψ Respectively as:

[0089]

[0090] Perform decoupling processing on Equation (1), and the UAV attitude dynamics model is rewritten as:

[0091]

[0092] In the formula, θ, ψ are the roll angle, pitch angle, and yaw angle of the UAV respectively; I x 、I y 、I z Are the moments of inertia of the aircraft about the x, y, and z axes respectively; Represents the roll angle angular acceleration, Represents the pitch angle angular acceleration, Represents the yaw angle angular acceleration, f θ 、f ψrespectively represent the total disturbances including internal and external disturbances on the roll channel, pitch channel, and yaw channel; \(l\) is the length of the lever arm; \(\tau\) θ and \(\tau\) ψ are the roll moment, pitch moment, and yaw moment respectively; \(k\) θ and \(k\) ψ are the drag coefficients; \(d\) θ and \(d\) ψ are the disturbance terms caused by external factors.

[0093] Specifically, in this embodiment, based on the UAV attitude dynamics model, an attitude controller based on improved LQR-LADRC is constructed; based on the attitude controller, the system control quantity capable of driving the UAV attitude to adjust to the desired state is calculated;

[0094] The desired angular velocity is obtained by using P control for the attitude angle error, and the angular velocity error is introduced. Taking the attitude angle error and angular velocity error as state variables, an optimal control law is designed to make the attitude angle error and angular velocity error quickly converge to 0; among them, the optimal control law design of the roll angle is as follows:

[0095] Using P control for the error between the desired roll angle and the actual roll angle to obtain the desired angular velocity, the desired roll angular velocity is:

[0096]

[0097] In the formula, is the proportional coefficient of P control, is the desired attitude angle, output by the UAV position controller; the following simplification is made: since the output by the position controller each time is a fixed value, the change rate of the desired roll angle can be taken as 0, that is:

[0098]

[0099] Taking the roll angle error and roll angular velocity error as state variables and

[0100]

[0101] Taking the control quantity as:

[0102]

[0103] From equations (5), (6), (7), and (8), the state space equation of the roll angle channel can be rewritten as:

[0104]

[0105] That is, the roll angle channel system matrix is:

[0106]

[0107] The control matrix is:

[0108]

[0109] Take the state variable weighting matrix as:

[0110]

[0111] Take the control variable weighting matrix as:

[0112]

[0113] By solving the Riccati equation, calculate the matrix

[0114]

[0115] Based on the matrix Calculate the roll channel feedback gain matrix

[0116]

[0117] Obtain the optimal control law for the roll angle channel as:

[0118]

[0119] Where, is the optimal control law for the roll angle channel, and are the elements of the gain matrix.

[0120] The dynamic characteristics of the quadrotor UAV in the roll channel and the pitch channel are similar. Therefore, the controller parameters for the roll angle and the pitch angle are designed in a general way. The optimal control law for the pitch angle channel is:

[0121]

[0122] The design process of the optimal control law for the yaw angle channel is the same as that for the roll angle. The optimal control law is:

[0123]

[0124] Where, u 0ψ is the optimal control law for the pitch angle channel, is the yaw channel feedback gain matrix, x1ψ = ψ d - ψ is the yaw angle error state variable, is the yaw angular velocity error state variable, is the yaw angular velocity.

[0125] Furthermore, in the S2, a second-order extended state observer (LESO) is introduced to estimate the total disturbance and perform feedforward compensation, and combined with the optimal control law, the system control quantity is obtained; among them, the roll angle LESO is:

[0126]

[0127] Among them, are the estimated values of the roll angle roll angular velocity total disturbance of the roll channel respectively, are their rates of change, is the observer bandwidth of the roll angle and pitch angle;

[0128] The pitch angle LESO is:

[0129]

[0130] Among them, z 1θ , z 2θ , z 3θ are the estimated values of the pitch angle θ, pitch angular velocity total disturbance f of the pitch channel θ respectively, are their rates of change, is the observer bandwidth of the roll angle and pitch angle;

[0131] The yaw angle LESO is:

[0132]

[0133] Among them, z 1ψ , z 2ψ , z 3ψ are the estimated values of the yaw angle ψ, yaw angular velocity total disturbance f of the yaw channel ψ respectively, are their rates of change, w oψ is the observer bandwidth of the yaw angle. b θ , b ψ The values are:

[0134]

[0135] Combined with the optimal control law, the system control quantity is:

[0136]

[0137] Among them, represents the control quantity of the roll channel system, u θ represents the control quantity of the pitch channel system, u ψ represents the control quantity of the yaw channel system, represents the disturbance compensation factor of the roll channel, b θ represents the disturbance compensation factor of the pitch channel, b ψ represents the disturbance compensation factor of the yaw channel.

[0138] Next, taking the roll angle as an example in this embodiment, four controllers, namely improved LQR-LADRC, LQR-LADRC, improved LQR, and LQR, are designed, and the four controllers are simulated and compared in the Matlab / Simulink environment; among them, for the optimal control rates in the LQR-LADRC and LQR controllers, the roll angle error and the roll angle velocity are taken as state variables for design; the relevant parameters are shown in Tables 1 and 2:

[0139] Table 1 Parameters of the quadrotor UAV model

[0140]

[0141] Table 2 Controller parameters

[0142]

[0143] As Figure 3 shown is the response diagram of each controller to a 15° roll angle step signal in an environment without external disturbances; there is no obvious difference whether to introduce LESO without external disturbances; the improved LQR-LADRC and the improved LQR converge to the expected value at about 0.84 s without overshoot; the LQR reaches stability at about 0.55 s with an overshoot of about 0.33%; the LQR-LADRC reaches stability at about 0.81 s with an overshoot of about 0.33%; the improved LQR-LADRC and the improved LQR have a slightly slower settling time, but they have a faster rising speed in the initial stage of the response, that is, they can respond at a faster speed. For devices with high real-time requirements such as quadrotor UAVs, the improved LQR-LADRC and the improved LQR have advantages.

[0144] Figure 4 The figure shows the response diagram of each controller to a 15° roll angle step signal under a sinusoidal disturbance, and the applied sinusoidal disturbance is Table 3 shows the maximum steady-state error of each controller.

[0145] Table 3 Maximum steady-state error of each controller

[0146]

[0147] The improved LQR can respond more quickly to the roll angle error, thereby reducing the steady-state error of the system under disturbances; introducing the LESO can improve the disturbance rejection ability of the system and further reduce the steady-state error; compared with the other three controllers, the improved LQR-LADRC controller of the present invention has the smallest maximum steady-state error value and has a better disturbance rejection effect.

[0148] The embodiments described above are only descriptions of the preferred embodiments of the present invention and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. An anti-wind control method for unmanned aerial vehicles based on improved LQR-LADRC, characterized in that include: Construct a UAV attitude dynamics model with disturbance terms; Based on the UAV attitude dynamics model, an attitude controller based on an improved LQR-LADRC is constructed; wherein the improved LQR-LADRC attitude controller includes: calculating the desired angular velocity based on the attitude angle error, designing the optimal control rate based on the attitude angle error and the angular velocity error, and introducing a second-order extended state observer; Based on the improved LQR-LADRC attitude controller, a system control variable capable of driving the UAV attitude to adjust to a desired state is calculated; The UAV wind resistance control is performed based on the system control quantity.

2. The UAV wind resistance control method based on improved LQR-LADRC according to claim 1 is characterized in that: The UAV attitude dynamics model is: Among them, θ and ψ are the roll angle, pitch angle, and yaw angle of the UAV respectively, and I x , I y , I z are the moments of inertia about the x, y, and z axes of the airframe respectively, represents the roll angle angular acceleration, represents the pitch angle angular acceleration, represents the yaw angle angular acceleration, f θ , f ψ represent the total disturbances including internal and external disturbances on the roll channel, pitch channel, and yaw channel respectively.

3. The method for controlling the wind resistance of an unmanned aerial vehicle based on improved LQR-LADRC according to claim 1, characterized in that Based on the improved LQR-LADRC attitude controller, the system control variables that can drive the UAV attitude to adjust to the desired state are calculated as follows: The desired angular velocity is obtained by applying P control to the attitude angle error, and the angular velocity error is introduced. The attitude angle error and angular velocity error are used as state variables, and the optimal control rate is designed to make the attitude angle error and angular velocity error converge quickly to 0. A second-order extended state observer is introduced to estimate the total disturbance and perform feedforward compensation, and combined with the optimal control rate, the system control quantity is obtained.

4. The method for controlling the wind resistance of an unmanned aerial vehicle based on the improved LQR-LADRC according to claim 3, wherein The optimal control rate includes: the optimal control rate of the roll angle channel, the optimal control rate of the pitch angle channel, and the optimal control rate of the yaw angle channel.

5. The method for controlling the wind resistance of an unmanned aerial vehicle based on the improved LQR-LADRC according to claim 4, characterized in that, Designing the optimal control rate includes: Design the optimal control rate of the roll angle channel, including: The error between the desired roll angle and the actual roll angle is controlled by P to obtain the desired roll angular velocity; the desired roll angular velocity is as follows: wherein, is the desired attitude angle, output by the UAV position controller, is the proportional coefficient of P control; Based on the output of the position controller is a constant value, then the expected roll angle change rate is taken as 0, that is: Take the roll angle error and the roll angular velocity error as state variables and wherein, is the roll angular velocity; Obtain the control quantity as follows: Based on the desired roll angular velocity The desired roll angular rate is 0, the roll angle error, the roll angular velocity error, and the control quantity The state space equation for obtaining the roll angle channel is as follows: That is, the roll angle channel system matrix is: The control matrix is: The state variable weight matrix is taken as: The control variable weight matrix is taken as: By solving the Riccati equation, calculate the matrix Based on the matrix Calculate the roll channel feedback gain matrix The optimal control rate of the roll angle channel is obtained as: in, is the optimal control rate of the roll angle channel, and is the gain matrix element.

6. The method for controlling the wind resistance of an unmanned aerial vehicle based on the improved LQR-LADRC according to claim 5, characterized in that, Designing the optimal control rate of the pitch angle channel includes: The controller parameters of the pitch angle channel are designed in common with the roll angle channel; the optimal control rate of the pitch angle channel is: Among them, u 0θ is the optimal control rate of the pitch angle channel.

7. The method for controlling the wind resistance of an unmanned aerial vehicle based on the improved LQR-LADRC according to claim 5, characterized in that Designing the optimal control rate of the yaw angle channel includes: The design process of the optimal control rate of the yaw angle channel is consistent with the design process of the optimal control rate of the roll angle channel; the optimal control rate of the yaw angle channel is: where, u 0ψ is the optimal control rate of the pitch angle channel, is the feedback gain matrix of the yaw channel, x 1ψ = ψ d -ψ is the yaw angle error state variable, is the yaw angular velocity error state variable, is the yaw angular velocity.

8. The method for controlling the wind resistance of an unmanned aerial vehicle based on the improved LQR-LADRC according to claim 3, wherein The second-order extended state observer includes: a second-order extended state observer for roll angle, a second-order extended state observer for pitch angle, and a second-order extended state observer for yaw angle; The second-order extended state observer of the roll angle is: Among them, are the estimated values of the roll angle roll angle angular velocity total disturbance of the roll channel respectively, are their rates of change respectively, is the observer bandwidth of the roll angle and pitch angle; The second-order extended state observer of the pitch angle is: Among them, z 1θ , z 2θ , z 3θ are the estimated values of the pitch angle θ, the pitch angular velocity and the total disturbance f θ of the pitch channel respectively, are their rates of change respectively, and are the bandwidths of the roll angle and pitch angle observers; The second-order extended state observer of the yaw angle is: Among them, z 1ψ , z 2ψ , z 3ψ are the estimated values of the yaw angle ψ, the yaw angular velocity and the total disturbance f ψ of the yaw channel respectively, are their rates of change respectively, and w oψ is the bandwidth of the yaw angle observer.

9. The method for controlling the wind resistance of an unmanned aerial vehicle based on improved LQR-LADRC according to claim 1, wherein, The system control quantity is: in, Indicates the control quantity of the roll channel system, u θ Indicates the control quantity of the pitch channel system, u ψ represents the control quantity of the yaw channel system, represents the roll channel disturbance compensation factor, b θ represents the pitch channel disturbance compensation factor, b ψ Represents the yaw channel disturbance compensation factor.

Images