A control method for quadrotor forward flight decoupling based on improved active disturbance rejection

Abstract

The application discloses a kind of based on improved self-disturbance four-rotor front flight decoupling control method, and the attitude decoupling of high-speed front flight dynamics of four-rotor is carried out by improved self-disturbance controller, and attitude channel control instruction is obtained;The improved self-disturbance controller is a kind of ADRC controller, which can take the coupling term in the decoupling channel as one of the total disturbance of the model, and compensate to control instruction.This application provides a kind of based on improved self-disturbance four-rotor front flight decoupling control method, solves the problem of slow convergence speed and poor stability of four-rotor front flight state, and the overall steady-state error is relatively small, the control accuracy of input signal is high, and has good robustness.

Description

Technical Field

[0001] This invention relates to a control method for decoupling quadrotor forward flight based on improved active disturbance rejection, belonging to the field of aircraft control technology. Background Technology

[0002] Because quadcopter drone systems are inherently underactuated and strongly coupled, they are highly susceptible to external environmental interference such as gusts and ground effects.

[0003] Existing technologies generally employ control methods such as active disturbance rejection control or fast terminal sliding mode control to solve this problem. Based on the decoupling design with frequency domain margin compensation, this method has achieved good results in the hovering state of a quadcopter UAV.

[0004] However, when quadcopter UAVs perform complex maneuvers in forward flight mode, the strong coupling and mutual interference between control channels become more pronounced. Traditional control methods have certain limitations and are easily affected by interference, resulting in slow convergence speed and poor stability.

[0005] Therefore, it is necessary to conduct further research on the forward flight state of quadcopters in order to solve the above problems. Summary of the Invention

[0006] To overcome the above problems, the inventors conducted in-depth research and found that in the forward flight state of a quadrotor UAV, there is a problem of coupling interference between state variables and control commands between pitch, roll, yaw and vertical channels. A control method based on improved active disturbance rejection for quadrotor forward flight decoupling is proposed. By improving the active disturbance rejection controller, the attitude decoupling of the quadrotor's high-speed forward flight dynamics is performed to obtain attitude channel control commands.

[0007] The attitude channels include roll, pitch, yaw, and vertical channels;

[0008] The improved active disturbance rejection controller is an ADRC controller that can treat the coupling terms of the decoupled channels as one of the total disturbances in the model and compensate them into the control command.

[0009] Furthermore, the improved active disturbance rejection controller obtains the coupling terms of the channels by decoupling the attitude motion control into longitudinal and lateral sub-models and decomposing them into different attitude channels.

[0010] According to the present invention, the improved active disturbance rejection controller includes:

[0011] Tracking differentiator: used to acquire the input signal and its differential signal;

[0012] Extended State Observer: Estimates the angular motion state and total disturbance of the system after attitude decoupling;

[0013] Error compensation controller: suppresses and reduces disturbances based on the results of the tracking differentiator and the extended state observer;

[0014] The input signal is the desired value of the channel.

[0015] In a preferred embodiment, the extended state observer is provided with an adaptive gain that allows it to change according to the estimation error of the attitude motion information.

[0016] In a preferred embodiment, the extended state observer in the improved active disturbance rejection controller is configured as follows:

[0017]

[0018] Where j represents different channels, j = 1, 2, 3, 4 are the roll channel, pitch channel, yaw channel, and vertical channel, respectively; A j B represents the state matrix of different channel dynamics models. j The control matrix x represents the dynamic model of different channels. j δ represents the state variables of different channels. k These represent different channel control commands, with k = 1, 2, 3, and 4 representing the roll, pitch, yaw, and vertical channels, respectively; z1 and z2 represent the estimated values ​​of the channel system state variables, ρ i (t), i = 1, 2, 3, 4 represents the adaptive gain, sig 1-1 / γ The (ε) function represents the operational relation ||ε||. 1-1 / γ ·sign(ε), sig 1-2 / γ (ε) represents the operational relation ||ε|| 1-2 / γ ·sign(ε), where γ is a constant.

[0019] More preferably, the adaptive gain ρ i (t), i = 1, 2, 3, 4 are set as follows;

[0020]

[0021] Where λ>0 is a constant, and c1, c2, c3, and c4 are settable coefficients.

[0022] Furthermore, the settable coefficients in the adaptive gain satisfy:

[0023]

[0024] In a preferred embodiment, the tracking differentiator has a hyperbolic tangent function as the error synthesis control function.

[0025] In a preferred embodiment, the tracking differentiator is represented as:

[0026]

[0027] Where ε is the tracking error of the input signal v1, a0, b0, a1, b1, a2, b2 are positive constant coefficients, r represents the time scale factor, τ is a constant and 0 < τ < 1, u represents the system input; tanh(·) is the hyperbolic tangent function.

[0028] In a preferred embodiment, the hyperbolic tangent function is expressed as:

[0029]

[0030] The beneficial effects of this invention include:

[0031] (1) The improved extended state observer compensates for the disturbance caused by channel coupling in the control system, and solves the problems of slow convergence speed and poor stability of the quadrotor in forward flight state.

[0032] (2) It has good robustness and the quadcopter can quickly recover accurate tracking of attitude angle;

[0033] (3) The overall steady-state error is relatively small, and the control accuracy of the input signal is high;

[0034] (4) Fast attitude angular rate response. Attached Figure Description

[0035] Figure 1 A schematic flowchart of a control method for decoupling a quadrotor in forward flight based on an improved active disturbance rejection system according to a preferred embodiment of the present invention is shown.

[0036] Figure 2 The simulation results of Example 1 show the response of the attitude angle in the roll channel when the attitude control channel is excited by a sinusoidal signal;

[0037] Figure 3 The simulation results of Example 1 show the response of angular velocity in the roll channel when the attitude control channel is excited by a sinusoidal signal;

[0038] Figure 4 The simulation results of Example 1 show the response of the attitude angle in the pitch channel when the attitude control channel is excited by a sinusoidal signal;

[0039] Figure 5 The simulation results of Example 1 show the response of the angular rate in the pitch channel when the attitude control channel is excited by a sinusoidal signal;

[0040] Figure 6The simulation results of Example 1 show the response of the attitude angle in the yaw channel when the attitude control channel is excited by a sinusoidal signal;

[0041] Figure 7 The simulation results of Example 1 show the response of angular rate in the yaw channel when the attitude control channel is excited by a sinusoidal signal;

[0042] Figure 8 The simulation results of Example 1 show the response of the attitude angle in the roll channel after adding a disturbance;

[0043] Figure 9 The simulation results of Example 1 show the response of angular velocity in the roll channel after adding a disturbance;

[0044] Figure 10 The simulation results of Example 1 show the response of the attitude angle in the pitch channel after adding a disturbance;

[0045] Figure 11 The simulation results of Example 1 show the response of the angular rate in the pitch channel after adding a disturbance;

[0046] Figure 12 The simulation results of Example 1 show the response of the attitude angle in the yaw channel after adding a disturbance;

[0047] Figure 13 The simulation results of Example 1 show the response of angular rate in the yaw channel after adding disturbance. Detailed Implementation

[0048] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Through these descriptions, the features and advantages of the present invention will become clearer and more apparent.

[0049] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments. Although various aspects of embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated otherwise.

[0050] This invention provides a control method for decoupling quadrotor forward flight based on improved active disturbance rejection. By improving the active disturbance rejection controller, the attitude decoupling of the quadrotor's high-speed forward flight dynamics is performed to obtain attitude channel control commands.

[0051] The attitude channels include roll, pitch, yaw, and vertical channels.

[0052] Furthermore, the inventors discovered that there is a problem of coupling interference between state variables and control commands in roll, pitch, yaw, and vertical channels, especially in the forward flight mode of quadcopter UAVs, where the interference is more significant.

[0053] In this invention, the control concept of ADRC controller is applied to the design of multivariable coupled systems, and the traditional ADRC controller is improved so that the improved active disturbance rejection controller can treat the coupling terms of the channels in the decoupling process as one of the total disturbances in the model and compensate it into the control command, thereby solving the strong coupling and mutual interference between channels.

[0054] Furthermore, the improved active disturbance rejection controller obtains the coupling terms of the channels by decoupling the attitude motion control into longitudinal and lateral sub-models and decomposing them into different attitude channels.

[0055] Specifically, the vertical sub-model can be represented as:

[0056]

[0057] Where u represents the UAV's x-axis velocity, w represents the z-axis velocity, q represents the pitch velocity, θ represents the pitch angle, and X u Let represent the derivative of the force acting on the quadrotor in the u direction with respect to u in the body coordinate system, w0 represent the state value of w at the equilibrium point, g represent the acceleration due to gravity, θ0 represent the state value of θ at the equilibrium point, u0 represent the state value of u at the equilibrium point, and M represent the state value of u at the equilibrium point. u M represents the derivative of the pitch torque with respect to the velocity along the UAV's x-axis. w X represents the derivative of the pitch moment with respect to the UAV's z-axis velocity. lon Z represents the derivatives of the aerodynamic forces acting on the aircraft in the x and z directions with respect to the pitch control command in the body coordinate system. col M represents the derivative of the aerodynamic force in the z-direction with respect to the vertical control command in the body coordinate system. lon δ represents the derivative of the pitch moment with respect to the pitch control command. lon Indicates the system pitch command, δ col Indicates system altitude command,

[0058] The control input for the longitudinal sub-model is [δ lon ,δ col ] T The relevant state variables are [u, w, q, θ]. T .

[0059] The lateral sub-model can be represented as:

[0060]

[0061] Where v represents the y-axis velocity of the UAV, p represents the roll rate, r represents the yaw rate, φ represents the roll angle, and ψ represents the yaw angle. v Let w0 represent the derivative of the force acting on the quadrotor in the v-direction in the body coordinate system with respect to velocity, and L represent the state value of w at the equilibrium point.v N represents the derivative of the aerodynamic torque on the quadrotor in the v-direction in the body coordinate system with respect to velocity. r Y represents the damping moment coefficient in the r direction. lat L represents the derivative of the aerodynamic forces acting on the aircraft in the y and z directions with respect to the roll control command in the body coordinate system. lat N represents the derivative of the roll torque with respect to the roll control command. ped δ represents the derivative of the yaw moment with respect to the yaw control command. lat Indicates the rollover instruction, δ ped Indicates a yaw command.

[0062] The control input for the lateral sub-model is [δ lat ,δ ped ] T The relevant state variables are [v, p, r, φ, ψ]. T .

[0063] Furthermore, the longitudinal and lateral sub-models are decomposed into different attitude channels.

[0064] Specifically, when broken down to the roll channel, it is represented as:

[0065]

[0066]

[0067] Among them, [D v D p D φ ] T D represents the coupling term in the roll channel during the decoupling process. roll This represents the sum of model uncertainties and external disturbances that were not considered during the modeling process on the roll channel;

[0068] m represents the mass of the drone, Y w Y represents the derivative of the force on the y-axis of the drone with respect to w. q Y represents the derivative of the force on the y-axis of the drone with respect to q. r Let represent the derivative of the force on the y-axis of the UAV with respect to r, and let u0 represent the state variable of u at the equilibrium point. This represents the derivative of the force on the y-axis of the drone with respect to the roll command. δ represents the derivative of the UAV's roll torque with respect to the roll command. lat This indicates a rollover command; w0 represents the state of w at the equilibrium point; X col L represents the derivative of the aerodynamic force in the x-direction with respect to the vertical control command in the body coordinate system. u L represents the derivative of the torque along the x-axis of the UAV with respect to u. w L represents the derivative of the UAV's rolling torque with respect to w. qL represents the derivative of the UAV's rolling torque with respect to q. r I represents the derivative of the UAV's rolling torque with respect to r. xx I represents the moment of inertia of the UAV along the x-axis. yy L represents the moment of inertia of the UAV along the y-axis. ped φ represents the derivative of the roll torque with respect to the yaw control command, and φ0 represents the state value of φ at the equilibrium point.

[0069] Furthermore, in this invention, the improved active disturbance rejection controller incorporates the coupling term of the decoupled channel as a term in the total roll channel disturbance D1 of the model, expressed as:

[0070] D1 = [D v D p D φ ] T +D roll

[0071] The decoupled roll channel dynamics model then becomes:

[0072]

[0073] Where x1 represents the state variable of the channel, x1 = [v, p, φ] T A1 represents the state matrix of the roll channel dynamics model, and B1 represents the control matrix of the roll channel dynamics model.

[0074]

[0075] Decomposed into the pitch channel, it is represented as:

[0076]

[0077]

[0078] Among them, [D u D q D θ ] T D represents the coupling term in the pitch channel during the decoupling process. pitch This represents the sum of model uncertainties and external disturbances that were not considered during the modeling process on the pitch channel;

[0079] X v X represents the derivative of the force acting on the drone along the x-axis with respect to v. w X represents the derivative of the force on the drone's x-axis with respect to w. p X represents the derivative of the force on the drone's x-axis with respect to p. u X represents the derivative of the force acting on the drone along the x-axis with respect to u. q M represents the derivative of the force on the drone's x-axis with respect to q. uM represents the derivative of the pitching moment of the UAV with respect to u. v M represents the derivative of the pitching moment of the UAV with respect to v. p M represents the derivative of the pitch moment of the UAV with respect to p. w M represents the derivative of the pitch moment of the UAV with respect to w. r M represents the derivative of the pitch moment of the UAV with respect to r. col This represents the derivative of the drone's pitch moment with respect to the altitude command. This represents the derivative of the force on the UAV's x-axis with respect to the pitch command. δ represents the derivative of the UAV's pitch moment with respect to the pitch command. lon Indicates pitch command, I yy I represents the moment of inertia of the UAV along the y-axis. zz This represents the moment of inertia of the UAV along the z-axis.

[0080] The improved active disturbance rejection controller incorporates the coupling term of the decoupled channel as a term in the total pitch channel disturbance D2 of the model, expressed as:

[0081] D2=[D u D q D θ ] T +D pitch

[0082] The decoupled pitch channel dynamics model then transforms into:

[0083]

[0084] Where x2 represents the state variables of the channel, x2=[u,q,θ] T A2 represents the state matrix of the pitch channel dynamics model, and B2 represents the control matrix of the pitch channel dynamics model.

[0085]

[0086] Decomposed into yaw channels, it is represented as:

[0087]

[0088]

[0089] Among them, [D r D ψ ] T D represents the coupling term in the yaw channel during the decoupling process. yaw This represents the sum of model uncertainties and external disturbances that were not considered during the modeling process on the yaw channel;

[0090] N r N represents the derivative of the yaw moment with respect to r.u N represents the derivative of the yaw moment with respect to u. v N represents the derivative of the yaw moment with respect to v. w N represents the derivative of the yaw moment with respect to w. p N represents the derivative of the yaw moment with respect to p. q N represents the derivative of the yaw moment with respect to q. lat This represents the derivative of the yaw moment with respect to the roll control command. δ represents the derivative of the yaw moment with respect to the yaw control command. lat Indicates the rollover instruction, δ ped This indicates a yaw command.

[0091] The improved active disturbance rejection controller incorporates the coupling term of the decoupled middle channel as a term in the total yaw channel disturbance D3 of the model, expressed as:

[0092] D3 = [D r D ψ ] T +D yaw

[0093] The decoupled yaw channel dynamics model then becomes:

[0094]

[0095] Where x3 represents the state variable of the channel, x3 = [r, ψ] T A3 represents the state matrix of the yaw channel dynamics model, and B3 represents the control matrix of the yaw channel dynamics model.

[0096]

[0097] Decomposed into the vertical channel, it can be represented as:

[0098]

[0099] Among them, D m D represents the coupling term in the vertical channel during the decoupling process. vertical This represents the sum of model uncertainties and external disturbances that were not considered during the modeling process on the vertical channel;

[0100] Z w This represents the derivative of the stress along the z-axis with respect to w. δ represents the derivative of the stress along the z-axis with respect to the height command. col Indicates altitude command.

[0101] The improved active disturbance rejection controller incorporates the coupling term of the decoupled channel as a term in the total vertical channel disturbance D4 of the model, expressed as:

[0102] D4 = Dm +D vertical

[0103] The decoupled vertical channel dynamics model then transforms into:

[0104]

[0105] Where x4 represents the state variable of the channel, x4 = w, A4 represents the state matrix of the vertical channel dynamics model, and B4 represents the control matrix of the vertical channel dynamics model.

[0106]

[0107] Furthermore, in this invention, the improved active disturbance rejection controller is the same as the conventional ADRC controller, both comprising:

[0108] Tracking differentiator: used to acquire the input signal and its differential signal;

[0109] Extended State Observer: Estimates the angular motion state and total disturbance of the system after attitude decoupling;

[0110] Error compensation controller: suppresses and reduces disturbances based on the results of the tracking differentiator and the extended state observer;

[0111] The input signal is the desired value of the channel.

[0112] Unlike traditional ADRC controllers, the improved active disturbance rejection controller incorporates an adaptive gain in the extended state observer, enabling it to adapt to changes in the estimation error of attitude motion information.

[0113] Furthermore, the extended state observer in the improved active disturbance rejection controller is configured as follows:

[0114]

[0115] Where j represents different channels, j = 1, 2, 3, 4 are the roll channel, pitch channel, yaw channel, and vertical channel, respectively; A j B represents the state matrix of different channel dynamics models. j The control matrix x represents the dynamic model of different channels. j δ represents the state variables of different channels. k These represent different channel control commands, with k = 1, 2, 3, and 4 representing the roll, pitch, yaw, and vertical channels, respectively; z1 and z2 represent the estimated values ​​of the channel system state variables, ρ i (t), i = 1, 2, 3, 4 represents the adaptive gain, sig 1-1 / γ The (ε) function represents the operational relation ||ε||. 1-1 / γ·sign(ε), sig 1-2 / γ (ε) represents the operational relation ||ε|| 1-2 / γ ·sign(ε), where γ is a constant.

[0116] In a preferred embodiment, γ > 2.

[0117] More preferably, the adaptive gain ρ i (t), i = 1, 2, 3, 4 are set as follows;

[0118]

[0119] Where λ>0 is a constant, λ is the main rate of change affecting the adaptive gain, and c1, c2, c3, and c4 are settable coefficients.

[0120] More preferably, the settable coefficients in the adaptive gain satisfy:

[0121]

[0122] This ensures that the error of the total disturbance observed by the observer can converge to 0 within a finite time.

[0123] Furthermore, in this invention, the tracking differentiator of the traditional ADRC controller is improved by setting a hyperbolic tangent function as the error synthesis control function in the tracking differentiator, resulting in a tracking differentiator containing a nonlinear hybrid form.

[0124] Furthermore, power terms and terminal attractors are incorporated into the tracking differentiator, thereby improving its performance in areas such as fast convergence of dynamic response, chatter avoidance, and low-frequency filtering.

[0125] Furthermore, in this invention, a time-scale transformation is added to the tracking differentiator, and the final tracking differentiator can be expressed as:

[0126]

[0127] Where ε is the tracking error of the input signal v1, a0, b0, a1, b1, a2, b2 are positive constant coefficients, r represents the time scale factor, τ is a constant, and 0 < τ < 1, u 入 This represents the system input; tanh(·) is the hyperbolic tangent function.

[0128] More preferably, the hyperbolic tangent function is:

[0129]

[0130] Furthermore, by selecting appropriate coefficients, the following equation is satisfied for any time constant T and a bounded integrable input signal u(t):

[0131]

[0132] Example

[0133] Example 1

[0134] Simulation experiments were conducted to decouple the attitude of the quadrotor's high-speed forward flight dynamics by improving the active disturbance rejection controller (ADRC) to obtain attitude channel control commands. The improved ADRC is an ADRC controller that can treat the coupling terms of the decoupled channels as one of the total disturbances in the model and compensate them into the control commands.

[0135] In the simulation experiment, the improved active disturbance rejection controller is an ADRC controller that can treat the coupling terms of the decoupled channels as one of the total disturbances in the model and compensate them into the control command.

[0136] Furthermore, the extended state observer in the improved active disturbance rejection controller is set as follows:

[0137]

[0138] Adaptive gain ρ i (t), i = 1, 2, 3, 4 are set as follows:

[0139]

[0140] Where λ takes the value of 5, and c1, c2, c3, and c4 take the values ​​of 0.8, 0.5, 0.1, and 0.3, respectively.

[0141] The tracking differentiator is represented as:

[0142]

[0143] The values ​​of a0, b0, a1, b1, a2, and b2 are 0.25, 0.8, 0.005, 0.2, 0.005, and 0.2, respectively. The value of r is 10, and the value of τ is 0.5.

[0144] The hyperbolic tangent function is:

[0145]

[0146] Sine signals [0.175sin(0.4πt), 0.175sin(0.4πt), 0.175sin(0.2πt)] were used respectively. T To excite each attitude control channel, we obtain its attitude response curve without considering disturbances, as shown below. Figure 2-7As shown, where, Figure 2 , 3 The response to attitude angles and angular rates in the roll channel; Figure 4 , 5 The response to attitude angles and angular rates in the pitch channel; Figure 6 , 7 The response to attitude angles and angular rates in the yaw channel.

[0147] from Figure 2-7 The results show that the quadrotor has good attitude tracking performance and fast dynamic response. It is not affected by the tracking differentiator and the extended state observer. The lag of the command is consistent with the time delay of the transition process arranged by the tracking differentiator and the overall system time constant. At the same time, its overall stability is high. No overshoot or oscillation phenomenon is found in the attitude angle response. The change curve of the angular rate during the process is also relatively smooth. The analysis shows that the jump fluctuation in the initial segment is due to the sudden occurrence of the quadrotor's attitude motion in the simulation. The overall attitude angular rate response results and maneuver range can ensure the stability of the control system.

[0148] A disturbance is introduced at time 5s in the simulation, and the attitude response curve of the improved active disturbance rejection controller considering the disturbance is obtained as follows: Figure 8-13 As shown, where Figure 8 , 9 The response to attitude angles and angular rates in the roll channel; Figure 10 , 11 The response to attitude angles and angular rates in the pitch channel; Figure 12 , 13 The response to attitude angles and angular rates in the yaw channel.

[0149] As shown in the figure, the improved active disturbance rejection controller exhibits good robustness. Under the premise of introducing a certain disturbance, the quadrotor can still quickly recover accurate tracking of the attitude angle. Furthermore, thanks to the finite-time convergence characteristic of the extended state observer, the overall steady-state error of the system is relatively small, and the control accuracy for sinusoidal input signals is high. On the other hand, the attitude angular rate response can quickly sense disturbances and make corresponding adjustments.

[0150] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," "outer," "front," and "rear," etc., indicate the orientation or positional relationship based on the orientation or positional relationship in the working state of this invention, and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0151] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0152] The present invention has been described above with reference to preferred embodiments; however, these embodiments are merely exemplary and illustrative. Various substitutions and modifications can be made to the present invention based on these embodiments, all of which fall within the scope of protection of the present invention.

Claims

1.A control method for quadrotor forward flight decoupling based on improved active disturbance rejection control, characterized in that, an attitude decoupling is performed on high-speed forward flight dynamics of a quadrotor by an improved active disturbance rejection controller to obtain an attitude channel control instruction; the attitude channel includes a roll, a pitch, a yaw and a vertical channel; the improved active disturbance rejection controller is an ADRC controller capable of compensating coupling terms in the decoupling channel to a control instruction as one of total disturbances of a model, the improved active disturbance rejection controller includes: a tracking differentiator for obtaining an input signal and a differential signal thereof; an extended state observer for estimating angular motion states and total disturbance quantities of a system after attitude decoupling; an error compensation controller for disturbance suppression and reduction according to results of the tracking differentiator and the extended state observer; the input signal is an expected value of a channel, an adaptive gain is provided in the extended state observer, so that it can change according to an estimation error of attitude motion information, the extended state observer in the improved active disturbance rejection controller is set as: wherein j represents different channels, j = 1, 2, 3, 4 respectively represent roll channel, pitch channel, yaw channel and vertical channel; A j represents state matrix of different channel dynamic model, B j represents control matrix of different channel dynamic model, x j represents state quantity of different channel, δ k represents different channel control instruction, k = 1, 2, 3, 4 respectively represent roll channel, pitch channel, yaw channel and vertical channel; z1, z2 represent estimated value of channel system state quantity, ε represents error between input expected signal and observation signal, ρ i (t), i = 1, 2, 3, 4 represent adaptive gain, sig 1-1 / γ (ε) function represents operation relationship ||ε|| 1-1 / γ ·sign(ε), sig 1-2 / γ (ε) represents operation relationship ||ε|| 1-2 / γ ·sign(ε), γ is constant. 2.The control method for quadrotor forward flight decoupling based on improved active disturbance rejection control according to claim 1, characterized in that, the improved active disturbance rejection controller obtains coupling terms of a channel by decoupling attitude motion control into a longitudinal sub-model and a lateral sub-model and decomposing them into different attitude channels. 3.The control method for quadrotor forward flight decoupling based on improved active disturbance rejection control according to claim 1, characterized in that, Adaptive gain p i are set as; wherein λ>0 is a constant, and c1, c2, c3 and c4 are settable coefficients. 4.The control method for quadrotor forward flight decoupling based on improved active disturbance rejection control according to claim 3, characterized in that, the settable coefficients in the adaptive gain satisfy: 5.The control method for quadrotor forward flight decoupling based on improved active disturbance rejection control according to claim 1, characterized in that, the tracking differentiator has a hyperbolic tangent function as an error synthesis control function. 6.The control method for quadrotor forward flight decoupling based on improved active disturbance rejection control according to claim 5, characterized in that, the tracking differentiator is expressed as: where ε is the tracking error of the input signal v1, a0, b0, a1, b1, a2, b2 are positive constant coefficients, r represents a time scale factor, τ is a constant, and 0 < τ < 1, u 入 denotes the system input; tanh(·) is the hyperbolic tangent function.

Images