CN108536019B - Self-adaptive control method of four-rotor aircraft based on hyperbolic tangent enhanced double-power approach law and fast terminal sliding mode surface - Google Patents

Abstract

A self-adaptive control method of a four-rotor aircraft based on hyperbolic tangent enhanced double-power approach law and a fast terminal sliding mode surface comprises the following steps: step 1, determining a transfer matrix from a body coordinate system based on a four-rotor aircraft to an inertial coordinate system based on the earth; step 2, analyzing a four-rotor aircraft dynamic model according to a Newton Euler formula; and 3, calculating a tracking error, and designing a controller according to the fast terminal sliding mode surface and the first derivative thereof. The method combines hyperbolic tangent enhanced double-power-order approach law sliding mode control and rapid terminal sliding mode control, can increase approach speed when the sliding mode surface is far away, reduces buffeting, improves the rapidity and robustness of a system, realizes rapid and stable control, realizes limited time control of tracking errors, and solves the problem that the tracking errors tend to 0 only when the time tends to be infinite in the traditional sliding mode surface. The interference boundary is estimated through self-adaptation, and the stability of the system is improved.

Description

Self-adaptive control method of four-rotor aircraft based on hyperbolic tangent enhanced double-power approach law and fast terminal sliding mode surface

Technical Field

The invention relates to a self-adaptive control method of a four-rotor aircraft based on a hyperbolic tangent enhanced double-power approach law and a fast terminal sliding mode surface.

Background

The four-rotor aircraft has attracted wide attention of domestic and foreign scholars and scientific research institutions due to the characteristics of simple structure, strong maneuverability and unique flight mode, and is rapidly one of the hotspots of international research at present. Compared with a fixed-wing aircraft, the rotary-wing aircraft can vertically lift, has low requirement on the environment, does not need a runway, reduces the cost and has great commercial value. The development of aircrafts makes many dangerous high-altitude operations easy and safe, so as to cause deterrence to other countries in the military aspect and greatly increase the working efficiency in the civil aspect. The four-rotor aircraft has strong flexibility, can realize rapid transition of motion and hovering at any time, and can be competent for more challenging flight tasks with less damage risk. In the field of scientific research, because a four-rotor aircraft has the dynamic characteristics of nonlinearity, under-actuation and strong coupling, researchers often use the four-rotor aircraft as an experimental carrier for theoretical research and method verification. An aircraft flight control system is built by relying on a small four-rotor aircraft to carry out high-performance motion control research on the aircraft, and the method is a hot research field of the current academic world.

The approach law sliding mode control has the characteristics that discontinuous control can be realized, the sliding mode is programmable and is not related to system parameters and disturbance. The approach law sliding mode not only can reasonably design the speed of reaching the sliding mode surface, reduce the time of the approach stage, improve the robustness of the system, but also can effectively weaken the buffeting problem in the sliding mode control. Currently, in the field of four-rotor control, approach law sliding mode control is less used. The enhanced approach law further accelerates the approach speed of the system to the sliding mode surface and simultaneously enables the buffeting to be smaller on the basis of the traditional approach law. Because the four-rotor aircraft can encounter external environment interference in flight, interference and compensation are carried out on the interference boundary through self-adaptation, and the stability of the system is improved.

Disclosure of Invention

In order to solve the problems that the traditional sliding mode surface can not realize limited time control, further accelerate the approaching speed of an approaching law and reduce buffeting, the method adopts the rapid terminal sliding mode control and the hyperbolic tangent enhanced double-power approaching law, avoids the singularity problem through the switching control idea, accelerates the approaching speed of a system to the sliding mode surface, reduces buffeting and realizes the limited time control. Meanwhile, interference and compensation are carried out on the interference boundary through self-adaption, and the stability of the system is improved.

The technical scheme proposed for solving the technical problems is as follows:

a self-adaptive control method of a four-rotor aircraft based on hyperbolic tangent enhanced double-power approach law and a fast terminal sliding mode surface comprises the following steps:

step 1, determining a transfer matrix from a body coordinate system based on a four-rotor aircraft to an inertial coordinate system based on the earth;

where psi, theta, phi are the yaw, pitch, roll angles of the aircraft, respectively, representing the angle of rotation of the aircraft about each axis of the inertial frame in sequence, T_ψTransition matrix, T, representing psi_θA transition matrix, T, representing theta_φA transition matrix representing phi;

step 2, analyzing a four-rotor aircraft dynamic model according to a Newton Euler formula, wherein the process is as follows:

2.1, the translation process comprises the following steps:

wherein x, y and z respectively represent the position of the four rotors under an inertial coordinate system, m represents the mass of the aircraft, g represents the gravity acceleration, mg represents the gravity borne by the four rotors, and the resultant force U generated by the four rotors_r；

2.2, the rotation process comprises the following steps:

wherein tau is_x、τ_y、τ_zRespectively representing the axial moment components, I, in the coordinate system of the machine body_xx、I_yy、I_zzRespectively representing the rotational inertia component of each axis on the coordinate system of the machine body, x represents cross product, w_p、w_q、w_rRespectively representing the attitude angular velocity components of each axis on the coordinate system of the body,

respectively representing the attitude angular acceleration components of all axes on the coordinate system of the machine body;

considering that the aircraft is in a low-speed flight or hovering state, consider

Then the formula (3) is represented as the formula (4) in the rotation process

2.3, connecting the vertical type (1), (2) and (4), and obtaining the dynamic model of the aircraft as shown in the formula (5)

Wherein

U_x、U_y、U_zThe input quantities of the three position controllers are respectively;

according to the formula (5), decoupling calculation is carried out on the position and posture relation, and the result is as follows:

wherein phi_dIs the desired signal value of phi, theta_dDesired signal value of theta, psi_dFor desired signal values of ψ, the arcsin function is an arcsine function and the arctan function is an arctangent function;

equation (5) can also be written in matrix form as follows:

wherein X₁＝[x,y,z,φ,θ,ψ]^T,

B(X)＝diag(1,1,1,b₁,b₂,b₃),U＝[U_x,U_y,U_z,τ_x,τ_y,τ_z]^T，

Step 3, calculating a tracking error, and designing a controller according to the fast terminal sliding mode surface and the first derivative thereof, wherein the process is as follows:

3.1, defining the tracking error and its first and second differentials:

e＝X₁-X_d (8)

wherein, X_d＝[x_d,y_d,z_d,φ_d,θ_d,ψ_d]^T，x_d,y_d,z_d,φ_d,θ_d,ψ_dThe conductive desired signals are x, y, z, phi, theta, psi, respectively,

i＝1，2,3,4,5,6，D_i，c_0i，c_1i，c_2i，e_i，

respectively corresponding ith element;

3.2, designing a quick terminal sliding mode surface:

wherein, sig^α(x)＝|x|^α·sign(x)，α₁＞α₂＞1，λ₁＞0，λ₂＞0；

Derivation of equation (11) yields:

order to

Formula (12) is simplified to formula (13)

But due to the presence of alpha (e)

When α (e) is 0 and β (e) is not equal to 0, the negative power term of (a) causes a singularity problem;

consider the method of handover control:

wherein q is_i(e),α_i(e),β_i(e) Q (e), alpha (e), beta (e) respectively,

combining formula (13) and formula (14) to obtain:

conjunctive formula (7), formula (10) and formula (15) yields:

3.3 design enhanced approach law

Wherein n(s) diag [ δ + (μ - δ) [1-tanh (γ | s) ]₁|^p)],…，δ+(μ-δ)[1-tanh(γ|s₆|^p)]]，N^-1(X) is the inverse of N (X), k₁＞0，k₂＞0，β₁＞1，0＜β₂Less than 1, delta is more than 0 and less than 1, gamma is more than 0, mu is more than 1, and p is a positive integer;

3.4, combined vertical (16) and formula (17), to obtain a controller

Wherein B is^-1(X) is the inverse of B (X),

respectively corresponding ith element;

the adaptive law is designed as follows:

step 4, property specification, the process is as follows:

when the system moves away from the sliding mode, | s | is large, N(s) approaches δ,

the approach speed of the system is accelerated; when the system approaches the sliding mode, | s | approaches 0, N(s) approaches μ,

the buffeting of the system is reduced.

The technical conception of the invention is as follows: aiming at a four-rotor aircraft system, a self-adaptive control method of the four-rotor aircraft based on hyperbolic tangent enhanced double-power approximation law and a fast terminal sliding mode surface is designed by combining double-power approximation law sliding mode control and fast terminal sliding mode control. The quick terminal sliding mode surface can realize the limited time control of the tracking error, and solves the problems that the time tends to be infinite and the error tends to be 0 in the traditional sliding mode surface. Based on the hyperbolic tangent enhanced approach law, the approach speed can be increased when the system is far away from the sliding mode surface, buffeting can be reduced, the rapidness and robustness of the system are improved, and rapid and stable control is realized. Meanwhile, interference and compensation are carried out on the interference boundary through self-adaption, and the stability of the system is improved.

The invention has the beneficial effects that: compared with the traditional double-power approach law sliding mode control, the method can increase the approach speed when the system is far away from the sliding mode, reduce buffeting and shorten the arrival time of the sliding mode, thereby enabling the system to realize stable convergence more quickly. In addition, the invention utilizes the quick terminal sliding mode, solves the problems that the time tends to be infinite and the error tends to be 0 in the traditional sliding mode surface, and realizes the limited time control. Meanwhile, interference and compensation are carried out on the interference boundary through self-adaption, and the stability of the system is improved.

Drawings

Fig. 1 is a schematic diagram of a position tracking effect of a four-rotor aircraft, wherein a dotted line represents "1" type enhanced double-power-law adaptive control under a linear sliding mode surface, and a dotted line represents hyperbolic tangent "mu" type enhanced double-power-law adaptive control under a fast terminal sliding mode surface.

Fig. 2 is a schematic diagram of an attitude tracking effect of a four-rotor aircraft, wherein a dotted line represents "1" type enhanced double-power-law adaptive control of a linear sliding mode surface, and a dotted line represents hyperbolic tangent "mu" type enhanced double-power-law adaptive control of a fast terminal sliding mode surface.

Fig. 3 is a schematic input diagram of a position controller for enhanced double power law adaptive control of a '1' type under a linear sliding mode surface of a four-rotor aircraft.

Fig. 4 is an input schematic diagram of a position controller for self-adaptive control of a sliding mode surface of a fast terminal of a four-rotor aircraft based on hyperbolic tangent 'mu' type enhanced double-power approach law.

Fig. 5 is an input schematic diagram of an attitude controller for "1" -type enhanced double-power-law adaptive control under a linear sliding mode surface of a four-rotor aircraft.

Fig. 6 is an input schematic diagram of an attitude controller for self-adaptive control of a sliding mode surface of a fast terminal of a four-rotor aircraft based on hyperbolic tangent 'mu' type enhanced double-power approach law.

Fig. 7 is a schematic diagram of local amplification of input of an attitude controller for "1" -type enhanced double-power approach law adaptive control under a linear sliding mode surface of a four-rotor aircraft.

Fig. 8 is a schematic diagram of local amplification of input of an attitude controller of a four-rotor aircraft fast terminal sliding mode surface based on hyperbolic tangent 'mu' type enhanced double-power approach law adaptive control.

Fig. 9 is an estimation of the boundary of the position disturbance of the sliding mode surface of the fast terminal of the four-rotor aircraft based on hyperbolic tangent 'mu' type enhanced double power approach law adaptive control.

Fig. 10 is an estimation of the boundary of attitude disturbance of a four-rotor aircraft fast terminal sliding mode surface based on hyperbolic tangent 'mu' type enhanced double power approach law adaptive control.

FIG. 11 is a control flow diagram of the present invention.

Detailed Description

The invention is further described below with reference to the accompanying drawings.

Referring to fig. 1 to 11, a self-adaptive control method of a quadrotor aircraft based on hyperbolic tangent enhanced double-power approach law and fast terminal sliding mode surface includes the following steps:

step 1, determining a transfer matrix from a body coordinate system based on a four-rotor aircraft to an inertial coordinate system based on the earth;

where psi, theta, phi are the yaw, pitch, roll angles of the aircraft, respectively, representing the angle of rotation of the aircraft about each axis of the inertial frame in sequence, T_ψTransition matrix, T, representing psi_θA transition matrix, T, representing theta_φA transition matrix representing phi;

step 2, analyzing a four-rotor aircraft dynamic model according to a Newton Euler formula, wherein the process is as follows:

2.1, the translation process comprises the following steps:

wherein x, y and z respectively represent the position of the four rotors under an inertial coordinate system, m represents the mass of the aircraft, g represents the gravity acceleration, mg represents the gravity borne by the four rotors, and the resultant force U generated by the four rotors_r；

2.2, the rotation process comprises the following steps:

wherein tau is_x、τ_y、τ_zRespectively representing the axial moment components, I, in the coordinate system of the machine body_xx、I_yy、I_zzRespectively representing the rotational inertia component of each axis on the coordinate system of the machine body, x represents cross product, w_p、w_q、w_rRespectively representing the attitude angular velocity components of each axis on the coordinate system of the body,

respectively representing the attitude angular acceleration components of all axes on the coordinate system of the machine body;

considering that the aircraft is in a low-speed flight or hovering state, consider

Then the formula (3) is represented as the formula (4) in the rotation process

2.3, connecting the vertical type (1), (2) and (4), and obtaining the dynamic model of the aircraft as shown in the formula (5)

Wherein

U_x、U_y、U_zThe input quantities of the three position controllers are respectively;

according to the formula (5), decoupling calculation is carried out on the position and posture relation, and the result is as follows:

wherein phi_dIs the desired signal value of phi, theta_dDesired signal value of theta, psi_dFor desired signal values of ψ, the arcsin function is an arcsine function and the arctan function is an arctangent function;

equation (5) can also be written in matrix form as follows:

wherein X₁＝[x,y,z,φ,θ,ψ]^T,

B(X)＝diag(1,1,1,b₁,b₂,b₃),U＝[U_x,U_y,U_z,τ_x,τ_y,τ_z]^T，

Step 3, calculating a tracking error, and designing a controller according to the fast terminal sliding mode surface and the first derivative thereof, wherein the process is as follows:

3.1, defining the tracking error and its first and second differentials:

e＝X₁-X_d (8)

wherein, X_d＝[x_d,y_d,z_d,φ_d,θ_d,ψ_d]^T，x_d,y_d,z_d,φ_d,θ_d,ψ_dThe conductive desired signals are x, y, z, phi, theta, psi, respectively,

i＝1，2,3,4,5,6，D_i，c_0i，c_1i，c_2i，e_i，

respectively corresponding ith element;

3.2, designing a quick terminal sliding mode surface:

wherein, sig^α(x)＝|x|^α·sign(x)，α₁＞α₂＞1，λ₁＞0，λ₂＞0；

Derivation of equation (11) yields:

order to

Formula (12) is simplified to formula (13)

But due to the presence of alpha (e)

When α (e) is 0 and β (e) is not equal to 0, the negative power term of (a) causes a singularity problem;

consider the method of handover control:

wherein qi (e), α_i(e),β_i(e) Q (e), alpha (e), beta (e) respectively,

combining formula (13) and formula (14) to obtain:

conjunctive formula (7), formula (10) and formula (15) yields:

3.3 design enhanced approach law

Wherein n(s) diag [ δ + (μ - δ) [1-tanh (γ | s) ]₁|^p)],…，δ+(μ-δ)[1-tanh(γ|s₆|^p)]]，N^-1(X) is the inverse of N (X), k₁＞0，k₂＞0，β₁＞1，0＜β₂Less than 1, delta is more than 0 and less than 1, gamma is more than 0, mu is more than 1, and p is a positive integer;

3.4, combined vertical (16) and formula (17), to obtain a controller

Wherein B is^-1(X) is the inverse of B (X),

respectively corresponding ith element;

the adaptive law is designed as follows:

step 4, property specification, the process is as follows:

when the system moves away from the sliding mode, | s | is large, N(s) approaches δ,

the approach speed of the system is accelerated; when the system approaches the sliding mode, | s | approaches 0, N(s) approaches μ,

the buffeting of the system is reduced.

In order to verify the effectiveness of the method, the invention provides a comparison between a sliding mode control method of a fast terminal sliding mode based on hyperbolic tangent mu-type enhanced double-power-law approach and a sliding mode control method of a linear sliding mode 1-type enhanced double-power-law approach, wherein the sliding mode control method comprises the following steps:

wherein the 1-type enhanced double power approximation law is

For more efficient comparison, all parameters of the system are consistent, i.e. x_d＝y_d＝z_d＝2、ψ_d0.5, fast terminal sliding mode surface parameter：λ₁＝0.5、λ₂＝2、α₁＝2、α₂1.1, epsilon 0.3, linear slip-form face: lambda [ alpha ]₁0.5, "μ" type enhanced proximity law parameter: k is a radical of₁＝1、k₂＝10、δ＝0.1、p＝1、γ＝1、μ＝10，β₁＝1.3、β₂0.7, enhanced approximation rule parameter of "1": k is a radical of₁＝1、k₂＝10、δ＝0.1、p＝1、γ＝1、β₁＝1.3、β₂0.7, adaptive initial value setting

p_0i＝p_1i＝p_2i＝0.1，ε_0i＝ε_1i＝ε_2i0.001, 1,2,3,4,5,6, interference parameter: d_x＝d_y＝d_z＝0.2sin(0.2t)、

Parameters of the four-rotor aircraft: 1.1 and I_xx＝1.22、I_yy＝1.22、I_zz2.2, g 9.81, sampling parameters: t is t_s＝0.007，N＝5000。

As can be seen from fig. 1 and 2, the adaptive control of the quadrotor aircraft based on the hyperbolic tangent enhanced double-power approach law and the fast terminal sliding mode surface can reach the expected position more quickly; with reference to fig. 3-8, the self-adaptive control of the quadrotor aircraft based on the hyperbolic tangent enhanced double-power approach law and the fast terminal sliding mode surface has smaller buffeting. Fig. 9 and 10 can see the effectiveness of the estimation of the adaptive epipolar.

In conclusion, the self-adaptive control of the four-rotor aircraft based on the hyperbolic tangent enhanced double-power approach law and the fast terminal sliding mode surface can reduce the buffeting, reduce the tracking time, improve the tracking performance and enable the system to enter stable convergence more quickly.

While the foregoing has described a preferred embodiment of the invention, it will be appreciated that the invention is not limited to the embodiment described, but is capable of numerous modifications without departing from the basic spirit and scope of the invention as set out in the appended claims.

Claims (1)

1. A self-adaptive control method of a four-rotor aircraft based on a hyperbolic tangent enhanced double-power approach law and a fast terminal sliding mode surface is characterized by comprising the following steps:

step 1, determining a transfer matrix from a body coordinate system based on a four-rotor aircraft to an inertial coordinate system based on the earth;

wherein psi, theta and phi are respectively the yaw angle, pitch angle and roll angle of the aircraft, and represent the angle of the aircraft sequentially rotating around each axis of the inertial coordinate system, and T_ψTransition matrix, T, representing psi_θA transition matrix, T, representing theta_φA transition matrix representing phi;

step 2, analyzing a four-rotor aircraft dynamic model according to a Newton Euler formula, wherein the process is as follows:

2.1, the translation process comprises the following steps:

wherein x, y and z respectively represent the position of the four rotors under an inertial coordinate system, m represents the mass of the aircraft, g represents the gravity acceleration, mg represents the gravity borne by the four rotors, and the resultant force U generated by the four rotors_r；

2.2, the rotation process comprises the following steps:

wherein tau is_x、τ_y、τ_zRespectively representing the axial moment components, I, in the coordinate system of the machine body_xx、I_yy、I_zzRespectively representing the rotational inertia component of each axis on the coordinate system of the machine body, x represents cross product, w_p、w_q、w_rRespectively representing the attitude angular velocity components of each axis on the coordinate system of the body,

respectively representing the attitude angular acceleration components of all axes on the coordinate system of the machine body;

considering that the aircraft is in a low-speed flight or hovering state, consider

Then the formula (3) is represented as the formula (4) in the rotation process

2.3, connecting the vertical type (1), (2) and (4), and obtaining the dynamic model of the aircraft as shown in the formula (5)

Wherein

U_x、U_y、U_zThe input quantities of the three position controllers are respectively;

according to the formula (5), decoupling calculation is carried out on the position and posture relation, and the result is as follows:

wherein phi_dIs the desired signal value of phi, theta_dDesired signal value of theta, psi_dFor desired signal values of ψ, the arcsin function is an arcsine function and the arctan function is an arctangent function;

considering further the case where interference exists, equation (5) can be written in a matrix form as follows:

wherein X₁＝[x,y,z,φ,θ,ψ]^T，

B(X)＝diag(1,1,1,b₁,b₂,b₃)，U＝[U_x,U_y,U_z,τ_x,τ_y,τ_z]^T，

Step 3, calculating a tracking error, and designing a controller according to the fast terminal sliding mode surface and the first derivative thereof, wherein the process is as follows:

3.1, defining the tracking error and its first and second differentials:

e＝X₁-X_d (8)

wherein, X_d＝[x_d,y_d,z_d,φ_d,θ_d,ψ_d]^T，x_d,y_d,z_d,φ_d,θ_d,ψ_dThe conductive desired signals are x, y, z, phi, theta, psi, respectively,

i＝1，2,3,4,5,6，D_i，c_0i，c_1i，c_2i，e_i，

respectively corresponding ith element;

3.2, designing a quick terminal sliding mode surface:

wherein, sig^α(x)＝|x|^α·sign(x)，α₁＞α₂＞1，λ₁＞0，λ₂＞0；

Derivation of equation (11) yields:

order to

Formula (12) is simplified to formula (13)

But because of

In existence of

When α (e) is 0 and β (e) is not equal to 0, the negative power term of (a) causes a singularity problem;

consider the method of handover control:

wherein q is_i(e),α_i(e),β_i(e) Q (e), α (e), β (e), i ═ 1,2,3,4,5, 6;

combining formula (13) and formula (14) to obtain:

conjunctive formula (7), formula (10) and formula (15) yields:

3.3 design enhanced approach law

Wherein n(s) diag [ δ + (μ - δ) [1-tanh (γ | s) ]₁|^p)],…，δ+(μ-δ)[1-tanh(γ|s₆|^p)]]，N^-1(X) is the inverse of N (X), k₁＞0，k₂＞0，β₁＞1，0＜β₂Less than 1, delta is more than 0 and less than 1, gamma is more than 0, mu is more than 1, and p is a positive integer;

3.4, combined vertical (16) and formula (17), to obtain a controller

Wherein B is^-1(X) is the inverse of B (X),

respectively corresponding ith element;

the adaptive law is designed as follows:

step 4, property specification, the process is as follows:

when the system moves away from the sliding mode, | s | is large, N(s) approaches δ,

the approach speed of the system is accelerated; when the system approaches the sliding mode, | s | approaches 0, N(s) approaches μ,

the buffeting of the system is reduced.

Images