KR20220076151A - Apparatus and method for controlling line of sight stabilization and a recording medium storing a computer program - Google Patents

Abstract

The present invention relates to an apparatus for controlling gaze stabilization, a control method, and a recording medium storing a computer program, as an apparatus for controlling gaze stabilization of a gimbal for moving a gaze toward an object by receiving a control signal, input of a command, and angular velocity of the gimbal and a tracking control unit configured to output a first synthesis control signal for compensating for friction within the gimbal according to a feedback input of a gaze error. and a stabilization control unit for outputting a second synthesis control signal for compensating for the effect of disturbance applied to the gimbal using a control signal and the first synthesis control signal; performance can be implemented.

Description

A recording medium storing a gaze stabilization control device, a control method, and a computer program

The present invention relates to a gaze stabilization control device, a control method, and a recording medium storing a computer program, and more particularly, to a gaze stabilization control device, a control method, and a computer program capable of realizing precise and stable performance against unspecified disturbances. It relates to the storage medium.

Line of sight (LOS) is a term that indicates the direction in which a designated sensor or camera is looking. represents a system that

A general gaze stabilization system implements a control target performance through a proportional-integrator-derivative (PID) control method or a PI-lead control method to which a lead compensator is applied. This control method is selectively applied according to the purpose of use of the gaze stabilization system, and, for example, when a relatively high-precision stabilization performance is required, a PI-lead control method capable of increasing the stability of the system is adopted.

On the other hand, the gaze stabilization system has variable friction characteristics depending on various internal/external factors such as temperature, assembly and processing tolerances, and lubrication conditions. For example, in the case of a military-purpose servo system, it should be possible to perform missions in a high/low temperature environment ranging from -32°C to +45°C in general. resulting in a change in the properties of friction of the gaze stabilization system, such as causing a change. In addition, machining errors such as assembly tolerances and deviations in preload amount of bearings that occur in the process of mass-producing the gaze stabilization system act as factors that change the friction characteristics of the system. Therefore, there is a need for a method for securing the accuracy of the gaze stabilization system by estimating the characteristics of the variable friction in the design or verification process of the control device and compensating for the estimated friction.

KR 10-2020-0052705 A

The present invention provides a gaze stabilization control apparatus capable of estimating and compensating for friction uncertainty in a condition in which irregular disturbance is applied, a control method, and a recording medium storing a computer program.

The present invention provides an apparatus for controlling gaze stabilization capable of realizing high-precision stabilization performance by compensating for friction that fluctuates according to internal and external environments of a system, a control method, and a recording medium storing a computer program.

The gaze stabilization control device according to an embodiment of the present invention is a gaze stabilization control device for a gimbal for receiving a control signal to move a gaze toward an object, and according to a command input, a feedback input of the angular velocity of the gimbal and a gaze error, the a tracking control unit for outputting a first synthesis control signal for compensating for friction in the gimbal; and a stabilization control unit configured to output a second synthesis control signal for compensating for an effect of disturbance applied to the gimbal by using a control signal and the first synthesis control signal.

The following control unit may include: a variable estimator for estimating the friction variable of the gimbal from the feedback input of the gimbal; and a friction observer for calculating the friction torque of the gimbal by using the command input and the estimated friction variable, and outputting the first synthesized control signal for compensating for the calculated friction torque.

The following control unit may estimate the friction variable of the gimbal using an adaptive control method for sinusoidal driving based on the sliding mode.

The tracking controller may output an updated first composite control signal according to the feedback input of the gimbal.

The stabilization control unit, the disturbance detector for extracting the disturbance signal output from the sensor installed in the gimbal; and a synthesizer configured to synthesize the first synthesized control signal, the control signal, and the disturbance signal to output a second synthesized control signal.

The stabilization control unit may compensate for the effect of disturbance applied to the gimbal in a PI-LEAD control method.

A gaze stabilization control method according to an embodiment of the present invention is a gaze stabilization control method of a gimbal for receiving a control signal to move a gaze toward an object, and according to a command input, a feedback input of the angular velocity of the gimbal and a gaze error, the outputting a first composite control signal for compensating for friction in the gimbal; and outputting a second combined control signal for compensating for an effect of disturbance applied to the gimbal by using a control signal and the first combined control signal.

The step of outputting the first composite control signal may include the step of estimating the friction variable of the gimbal using an adaptive control method for sinusoidal driving based on a sliding mode.

The step of outputting the first composite control signal may include defining a new error (r) obtained by combining a position error and a speed error for a desired position of the gimbal from a command input as an error equation; defining a state equation for estimating the internal state variable of the gimbal from the error equation; and deriving a first synthetic control signal including the friction variable of the gimbal from the state equation.

The error equation is defined as in Equation 1 below.

[Equation 1]

, θ는 실제 목적 위치, θ_d는 목적 위치, w는 실제 각속도, w_d는 목적 각속도이다. )(From here

, θ is the actual target position, θ _d is the target position, w is the actual angular velocity, and w _d is the target angular velocity. )

The state equation is defined as in Equation 2 below.

[Equation 2]

(where z ^{^'} is the estimated value of the rate of change dz/dt of the internal state variable, z^ is the estimated value of the internal state variable z, w is the angular velocity of the gimbal, μ ₀ is the design variable, a positive constant value, and sgn is the sign function ( sgn (positive number) = + sign return, sgn (negative number) = - sign return, g(w) denotes a function of angular velocity representing the Strivek effect.)

The process of deriving the first composite control signal is performed by the following Equation (3).

[Equation 3]

(Where u is the first synthesis control signal, k _d and λ are design variables, and when e=θ-θ _d is the position error with respect to the target line of sight (θ _d ), r is the combination of the position error and the speed error. one new error, σ ^{^0} , σ ^{^1} , σ ^{^2} is the estimated friction variable, z^ is the estimate of the gimbal's internal state variable, z^' is the estimate of the rate of change of the gimbal's internal state variable dz/dt, e' is the error between the target angular velocity and the actual angular velocity, θ _d ^'' is the estimate of the target position of the gimbal, and J is the moment of inertia of the system).

The process of deriving the first composite control signal may include calculating a friction variable that converges the new error r to 0 using Equations 4 to 6 below.

[Equation 4]

[Equation 5]

[Equation 6]

(Here, γ ₀ , γ ₁ , and γ ₂ are positive constant values as design variables)

The method may include updating each of the estimated friction variables σ ^{^ 0} , σ ^{^ 1} , and σ ^{^ 2} according to the feedback input of the gimbal.

The outputting of the second composite control signal may include: extracting a disturbance signal from a signal output from a sensor installed in the gimbal; and synthesizing the control signal, the first synthesized control signal, and the disturbance signal to derive a second synthesized control signal.

A process of outputting the second composite control signal uses a PI-LEAD control method.

The gaze stabilization control method according to an embodiment of the present invention may be applied to a recording medium storing a computer program for performing the above method.

According to an embodiment of the present invention, it is possible to improve the gaze stabilization performance of the system by estimating and compensating for friction that occurs according to the internal and external environments of the gimbal. In particular, it is possible to estimate a dynamic variable of friction, which is difficult to measure experimentally, and compensate for friction through the estimated dynamic variable. In addition, it is possible to stabilize the gaze stabilization performance in real time by compensating for friction for unpredictable disturbances that may occur during the use of the gimbal. Therefore, it is possible to obtain a high-definition and clear image by stably moving the gaze of the gimbal.

1 is a view schematically showing a two-axis gimbal of a gaze stabilization device to which a gaze stabilization control device according to an embodiment of the present invention is applied.
2 is a block diagram schematically showing an apparatus for controlling gaze stabilization according to an embodiment of the present invention.
3 is a view schematically showing a stabilization control unit of the gaze stabilization control apparatus according to an embodiment of the present invention.
4 is a conceptual diagram of a LuGre friction model applied in an embodiment of the present invention.
5 is a control flowchart of an apparatus for controlling gaze stabilization according to an embodiment of the present invention.
Fig. 6 is a control flow chart in the following control unit;

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in various different forms, and only the embodiments of the present invention allow the disclosure of the present invention to be complete, and the scope of the invention to those of ordinary skill in the art It is provided to fully inform In order to describe the invention in detail, the drawings may be exaggerated, and like reference numerals refer to like elements in the drawings.

1 is a view schematically showing a two-axis gimbal of a gaze stabilization device to which a gaze stabilization control device according to an embodiment of the present invention is applied.

The gimbal must be able to move its gaze within the required time within the field of FOR (Field of Regard) required by the system. rejection). The gimbal includes a body frame (B) for fixing to a moving platform, an azimuth gimbal connected to the body frame (B) to move in a turning direction, and a high-angle gimbal connected to an azimuth gimbal to perform a movement in the high-low direction. can The gimbal is designed to mechanically minimize the disturbance applied from the outside to the azimuth axis and the elevation axis of the azimuth gimbal. This gimbal is driven by a DC motor, and includes a resolver (not shown) and a gyro sensor (not shown) to control position and speed signals.

The gaze stabilization control apparatus according to an embodiment of the present invention may implement high-precision stabilization performance by estimating the uncertainty of internal and external friction of the gimbal under a condition in which disturbance is applied to the gimbal, and compensating for this.

2 is a block diagram schematically showing an apparatus for controlling gaze stabilization according to an embodiment of the present invention.

Referring to FIG. 2 , the gaze stabilization control apparatus 100 according to an embodiment of the present invention provides a first method for compensating for friction of the gimbal 10 according to a command input, an angular velocity of the gimbal 10, and a feedback input of a gaze error. Stabilization for outputting a second synthesis control signal for compensating for the effect of disturbance applied to the gimbal 10 using the tracking control unit 110 and the control signal and the first synthesis control signal for outputting the first synthesis control signal A control unit 120 may be included. In addition, the gaze stabilization control apparatus 100 according to an embodiment of the present invention may further include a main controller (not shown). That is, in the gaze stabilization apparatus 100 according to an embodiment of the present invention, when power is applied to the system, inspection of the gimbal, camera, thermal imaging equipment, etc. is made, and at this time, the tracking control unit 110 controls the angular velocity and gaze error of the gimbal. Upon receiving the feedback, the first composite control signal including the friction variable of the gimbal may be output and transmitted to the stabilization control unit 120 . After the system check is completed, a control signal or a third synthesized control signal using the control signal and the disturbance signal may be output through the stabilization control unit 120 to compensate for the effect of the disturbance applied to the gimbal. And, if necessary, the tracking control unit 110 may be operated to output the first composite signal including the friction variable of the gimbal and transmit it to the stabilization control unit 120 . In this case, the tracking control unit 110 and the stabilization control unit 120 may be controlled through the main control unit.

The tracking control unit 110 includes a variable estimator (not shown) for estimating the friction variable of the gimbal from the feedback input of the gimbal, and a friction observer (not shown) for outputting a first synthesis control signal using the command input and the estimated friction variable. ) may be included. In this case, the variable estimator may estimate the friction variable of the gimbal through the adaptive control method, and the friction observer may output the first synthetic control signal based on the sliding mode. Also, the tracking control unit 110 may update and output the first synthesis control signal according to the feedback input of the gimbal, and may transmit the updated first synthesis control signal to the stabilization control unit.

The stabilization control unit 120 includes a disturbance detector (not shown) for extracting a disturbance signal applied to the gimbal and a synthesizer (not shown) for synthesizing the first synthesized control signal, the control signal, and the disturbance signal to output a second synthesized control signal. ) may be included. In addition, the stabilization control unit 120 may include a disturbance transmitter for feeding back the disturbance signal extracted from the disturbance detector to the variable estimator of the tracking control unit 110 . The stabilization control unit 120 basically removes the disturbance in the PI-Lead control method, and when the first synthesis control signal is input through the tracking control unit 110, it outputs a second synthesis control signal in which the friction is removed or compensated for the gimbal. disturbance can be eliminated. Here, removing the disturbance of the gimbal means compensating for the effect exerted on the gimbal by the disturbance.

Hereinafter, a gaze stabilization control method according to an embodiment of the present invention will be described.

First, the Lagrange equation considering the kinetic energy and potential energy of the rigid bodies considered above to derive the equation of motion of the gimbal is as follows.

[Equation 1]

Here, T is kinetic energy, V is potential energy, qi is a generalized coordinate, and Qi is an external force.

Since the dynamic behavior of the rotating part due to friction is the main concern of this study, the following assumptions were made to simplify the derivation of the system's motion equation.

1) There is no movement of the body frame.

2) The center of gravity of the rotating body for each gimbal is the same with respect to the axis of rotation.

3) The gimbal system moves independently in the direction of turning and the direction of elevation.

For convenience, four coordinate systems are considered as shown in FIG. 1 . Here, i ^{^} _E , j ^{^} _E , and k ^{^} _E are inertial coordinate systems, and i ^{^} _B , j ^{^} _B , k ^{^} _B are coordinate systems fixed to the body frame. i ^{^} _p , j ^{^} _p , and k ^{^} _p are coordinate systems fixed to the azimuth gimbal, and i ^{^} _T , j ^{^} _T , k ^{^} _T are coordinate systems fixed to the elevation gimbal.

Since the azimuth gimbal can rotate only in the turning direction with respect to the body frame, only the rotation about the k ^{^} _E axis is possible. . (Here, R ^B _P means the coordinate transformation matrix between the body frame (B) and the azimuth gimbal frame (P). C means trigonometric function cos, and means trigonometric function sin)

[Equation 2]

In order to obtain the angular velocity of each coordinate system, an angular velocity component related to each coordinate system is defined as in Equation (3).

[Equation 3]

The angular velocity w _P of the azimuth gimbal is obtained as in Equation 4 below by adding the relative angular velocity to the angular velocity w _B of the body frame. (where w _P is the angular velocity of the azimuth gimbal and w _T is the angular velocity of the elevation gimbal)

[Equation 4]

축을 기준으로 상대적인 고저 방향 회전만 가능하며, 각도 ε 만큼 회전이 발생하였을 때 두 좌표계의 변환 행렬 및 각속도는 방위각 김벌의 경우과 동일한 방법으로 유도할 수 있다. (여기서 R^P _T 는 방위각 김벌 프레임(P)과 고각 김벌 프레임(T) 사이의 좌표변환 행렬을 의미함)The elevation gimbal is connected to the azimuth gimbal.

Only relative high-low rotation is possible based on the axis, and when rotation occurs by an angle ε, the transformation matrix and angular velocity of the two coordinate systems can be derived in the same way as in the case of azimuth gimbal. (Here, R ^P _T means the coordinate transformation matrix between the azimuth gimbal frame (P) and the elevation gimbal frame (T))

[Equation 5]

[Equation 6]

When the kinetic energy of each rigid body is obtained using the angular velocity obtained earlier, the kinetic energy (T) of the entire gimbal is expressed as in Equation 7 below.

[Equation 7]

(Here, T _P is the kinetic energy of the azimuth gimbal, T _T is the kinetic energy of the high-angle gimbal, Pi is the moment of inertia of the azimuth gimbal, and Ti is the moment of inertia of the high-angle gimbal)

If the masses of the body frame, azimuth gimbal, and elevation gimbal are m _B , m _P , m _T , and the position of the center of gravity is GB , G _P , _{G T} based on k ^{^} _E of the inertial coordinate system, the overall position of the stabilization device Energy is equal to the sum of potential energy due to gravity of each rigid body of the gimbal system as shown in Equation 8 below.

[Equation 8]

Generalized forces include time-varying forces acting externally and non-conservative forces that cannot be derived from potential energy. The general coordinate qi is the azimuth gimbal rotation angle η and the elevation gimbal rotation angle ε, and generalized forces corresponding to the general coordinates are expressed as Equations 9 and 10 below.

[Equation 9]

[Equation 10]

Here, u _p denotes an external torque that rotates the system in the turning direction, and u _T denotes an external torque that rotates the system in the high-low direction. In addition, T _Pf and T _TF represent the friction of the driving part in the turning direction and the vertical direction, respectively.

The motion equations of the azimuth gimbal and the elevation gimbal can be obtained by the Lagrange equation of Equation 1, and if simplified by the above assumption, the following Equation 11 can be derived for each driving unit.

[Equation 11]

Here, J is the moment of inertia, u is the input torque, and T _f is the friction torque.

On the other hand, the gaze stabilization system is basically controlled by the stabilization control unit 120 for stabilization, and the stabilization control unit 120 is composed of a low pass filter - a PI controller - a advance compensator. In this gaze stabilization system, basically, the PI-LEAD control method has a high possibility of divergence due to the vulnerability of the differentiator to high-frequency noise that appears in the traditional PID control method. By compensating for the delay problem using a forward compensator, the robustness and stability of the system can be improved.

3 is a diagram schematically illustrating a stabilization control unit of an apparatus for controlling gaze stabilization according to an embodiment of the present invention.

Referring to FIG. 3 , the stabilization control unit 120 may use a low-pass filter to remove high-frequency noise from a sensor signal to prevent frequency aliasing that may occur during analog-to-digital conversion. The error signal that has passed through the low-pass filter is subjected to PI operation and is a DC motor control signal through forward compensation. limit1 performs the Anti-Windup function by preventing signal saturation occurring in the integral operation, and limit2 limits the signal before delivering the final control signal of the controller to the digital-analog converter to match the digital-analog converter input range. . The frequency response characteristic of the PI-LEAD compensator of the stabilization control unit 120 mainly consists of the integral operation characteristic in the low-frequency region, and the low-frequency gain is increased by the integral operation to remove the friction effect, which is a non-linear element existing in the two-axis stabilization gimbal system, and error can be minimized. Integral operation can increase the low frequency gain, but there is a problem of time delay, so the system can be transferred to an unstable region, so this problem can be solved by applying advance compensation. As such, the PI-LEAD control method compensates the time delay caused by the integral operation by applying the advance compensation instead of the differential compensation, which is vulnerable to high-frequency noise in the PID control method, and improves the performance and stability by speeding up the system response.

From a microscopic point of view, the contact surfaces of two non-moving objects are very rough and can be seen as bonded to each other. 4 is a conceptual diagram of a LuGre friction model applied in an embodiment of the present invention. The LuGre friction model models the characteristics of such a contact surface with bristle, and the friction is equivalent to the stiffness and damping terms due to bending of these bristle. analysis model. In an embodiment of the present invention, a friction variable for compensating for friction of a gimbal is estimated, and a LuGre friction model is applied. The advantage of this LuGre friction model is that it can express the friction phenomenon as a function, including the static friction region, the low velocity region, and the relatively high velocity region. LuGre friction model is based on the dynamic characteristics of general friction, i.e., hysteresis, friction lag, change of break-away force according to external force, stick-slip, control system and It has the advantage of showing phenomena such as the limit cycle and the Stribeck effect in connection.

The LuGre friction model presents the average bending displacement z of the bristles as Equations 12 and 13 below.

[Equation 12]

[Equation 13]

where g(w) is a function of the angular velocity representing the Strevek effect.

In the LuGre model, the total friction torque (T _f ) can be expressed by the following Equation 14 in the form of adding the spring-damper force generated from the bending of the bristles and the viscous friction proportional to the relative angular velocity of the contact surface.

[Equation 14]

From Equations 12 to 14, it can be seen that the LuGre friction model can be implemented by six variables, ie, σ ₀ , σ ₁ , σ ₂ , f _s , f _c , and w _s .

In a steady state in which the speed of the system is constant, the internal state variable z _ss and the friction torque T _fss can be expressed by Equations 15 and 16, respectively.

[Equation 15]

[Equation 16]

Here, f _s is static friction, f _c is Coulomb friction, and w _s is a static parameter meaning the Strike speed.

On the other hand, for a stiction section having little speed, the state variable can be approximated as in Equation 17 below.

[Equation 17]

By substituting Equations 14 and 17 into Equation 11 and rearranging them, the equilibrium equation of the system can be linearized into a quadratic system such as Equation 18 below.

[Equation 18]

Here, σ ₀ and σ ₁ represent the stiffness and damping coefficient of the bristle, respectively, and σ ₂ is a dynamic parameter meaning the viscous friction coefficient.

In the speed control mode, the PI-LEAD control alone could not secure the robustness against the uncertainty of the friction variable, so a method of compensating for friction using a friction observer was reviewed.

Various methods have been studied and utilized in designing a friction observer for friction compensation. In the embodiment of the present invention, friction variables already estimated in a certain area are utilized while utilizing the existing PI-LEAD control for efficient design of the friction observer. consider one option.

First, the position error e and the velocity error e' with respect to the target gaze θ _d are defined as in Equation 19 below.

[Equation 19]

Assuming that all friction variables are known, Canudas de Wit proposes a friction observer for estimating internal state variables of friction as shown in Equations 20 and 21 below, and controls input and H(s) in Equation 22 below and 23.

[Equation 20]

[Equation 21]

[Equation 22]

[Equation 23]

At this time, if an appropriate H(s) is selected so that the transfer function G(s) becomes Positive Real, the observer error, T _f -T ^{^} _f , and the velocity error e' converge to zero. (Here, H(s) is an arbitrary value for which G(s) is a Positive Real, e is a position error, m is a mass, and s is an S-space variable by Laplace transform)

In a similar way, a method of controlling the gimbal using the tracking control unit 110 including a friction observer for friction torque and the stabilization control unit 120 of the PI-LEAD control method may be considered.

If the PI-LEAD control input ( U _PIL ), the speed to be followed is 0, is expressed in Laplace Domain as in Equation 24.

[Equation 24]

Observer error for friction torque, E _T (s)=T _f -T ^{^} _f , and the transfer function of the system is G(s)=(Js) ^-1 , the transfer function between E _T and W is Equation 25 below. (here, k _p is the proportional controller (P) gain, ki is the integral controller (I) gain, W is the angular velocity of the gimbal, and z and p are the pole and zero values of the lead compensator. )

[Equation 25]

Therefore, if the gain value of the PI controller and the pole and zero values of the LEAD controller are selected so that the transfer function becomes Positive Real in the same way as in Equation 23, a stable system design in which the error converges to 0 is possible. However, this is a possible result when all friction variables are known, and the stability of the system cannot be confirmed when the actual friction variables change.

However, if the uncertainty of the friction variable can be estimated, the stability of the system presented above can be secured.

Since the design of the variable estimator for estimating all variables is complicated and inefficient, the method of estimating only dynamic variables that have a large influence on the response characteristics of the system was considered.

Existing studies have suggested adaptive control methods for dynamic friction variables in various ways. In the present invention, the variable estimator is designed by applying the adaptive control method based on the sliding mode observer suggested by Wen-Fang Xie.

First, when the position error with respect to the target line of sight θ _d is e= θ-θ _d , a new error r combining the position error and the velocity error is defined as a sliding plane as shown in Equation 26 below.

[Equation 26]

, θ는 실제 목적 위치, θ_d는 목적 위치, w는 실제 각속도, w_d는 목적 각속도이다. 설계변수 λ는 0 이하이다(λ>0).)(From here

, θ is the actual target position, θ _d is the target position, w is the actual angular velocity, and w _d is the target angular velocity. The design variable λ is less than or equal to 0 (λ > 0).

By differentiating Equation 26 and substituting it in Equation 11, Equation 27 below is derived.

[Equation 27]

When the value obtained by dividing the absolute value of the angular velocity by the Stribeck function g(w) is a new function α(w), the friction observer for estimating the internal state variable z of friction is defined by Equation 28 below.

[Equation 28]

(where z ^{^'} is the estimated value of the rate of change dz/dt of the internal state variable, z^ is the estimated value of the internal state variable z, w is the angular velocity of the gimbal, μ ₀ is the design variable, a positive constant value, and sgn is the sign function ( sgn (positive number) = + sign return, sgn (negative number) = - sign return, g(w) denotes a function of angular velocity representing the Strivek effect.)

The equation for the error of the friction observer by subtracting Equation 12 from Equation 28 is as Equation 29 below.

[Equation 29]

Here, [*~]=[*^]-[*].

In this case, if the control input by the estimated variables is defined as in Equation 30, update laws for the estimated variables are derived as a function of the error r as shown in Equations 31 to 33 below.

[Equation 30]

(Here, u is a control input (first synthesized control signal), k _d , λ is a design variable, and when e=θ-θ _d is the position error with respect to the target line of sight (θ _d ), r is the position error and The new error combined with the velocity error, σ ^{^0} , σ ^{^1} , σ ^{^2} is the estimated friction variable, z^ is the estimated value of the gimbal's internal state variable, z^' is the rate of change of the gimbal's internal state variable dz/dt Estimated value, e' is the error between the target angular velocity and the actual angular velocity, θ _d ^'' is the estimated value of the target position of the gimbal, and J is the moment of inertia of the system)

[Equation 31]

[Equation 32]

[Equation 33]

Here, γ ₀ , γ ₁ , and γ ₂ are positive constants that determine the adaptation rate of the estimated variable.

The proposed variable estimator can be expressed as a block diagram as follows, and its stability is proven by the following theorem and proof. Here, the stability is demonstrated using the Lyapunov function, which is one of the methods to check the stability of a system described by a nonlinear equation. The Liafnov function is always positive (+) in the entire state space other than the equilibrium point of the control system, and the time derivative is always negative (-) in the entire space, and the generalized energy function of the system decreases with time. This means that the system is stable.

Theorem: The system of Equation 11 having friction given by Equations 12 and 14, the sliding mode observer of Equation 28, and the variable estimator of Equation 30 for estimating friction variables by the update rule of Equations 31 to 33 Considering that, the tracking error converges to zero and all closed-loop signals have bounded values.

Proof: The Liafnov function is defined as in Equation 34.

[Equation 34]

By differentiating both sides of Equation 34 and substituting Equations 27 to 33 to organize them, Equation 35 below can be derived.

[Equation 35]

In Equation 35, if the property of the arithmetic geometric mean inequality ab<1/2(a ² +b ² ) is used, an inequality as shown in Equation 36 below is established.

[Equation 36]

In Equation 36, if two conditions of Equations 37 and 38 below are satisfied, the inequality of Equation 39 is established.

[Equation 37]

[Equation 38]

[Equation 39]

Since the Liafnov function in Equation 34 is positive definite and V'≤0, V is bounded, which means that the observer error z ^~ , the tracking error r, and the friction variable estimation error σ ^~ ₀ , σ ^~ ₁ , and σ ^~B are all It means that it has a bounded value. In addition, the friction variable estimates σ ^{^} ₀ , σ ^{^} ₁ , σ ^{^} ₂ , the position error e, and the control input u are also bound.

Since r∈L ₂ , e∈L ₂ ∩L _∞ , the time t→∞ and according to e ^' ∈L ₂ , e is continuous and converges to zero. Since r∈L ₂ and θ _d , w _d , and w ^' _d ∈L _∞ , r ^' ∈L _∞ is determined from Equations 5 and 8. Using the fact that r∈L _∞ and r ^' ∈L _INF , when time approaches 0, r also approaches 0, and thus the velocity error e ^' also converges to 0.

Meanwhile, the gaze stabilization control method according to an embodiment of the present invention may be applied to a recording medium storing a computer program for performing the above method.

That is, the gaze stabilization control method according to an embodiment of the present invention may be implemented in the form of a computer-readable programming language code recorded on a recording medium. The computer-readable recording medium may be any data storage device readable by the computer and capable of storing data. For example, the computer-readable recording medium may be a ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical disk, hard disk drive, flash memory, solid state disk (SSD), or the like. In addition, the computer-readable code or program stored in the computer-readable recording medium may be transmitted through a network connected between computers.

As such, according to an embodiment of the present invention, it is possible to compensate the friction of the gimbal even in a state in which a disturbance is applied, so that it is possible to improve the performance of following a control command. That is, it is possible to improve the accuracy of the roll (roll), pitch (pitch) and yaw (yaw) driving of the gimbal for controlling the gaze toward the object.

In the above, preferred embodiments of the present invention have been described and illustrated using specific terms, but such terms are only for clearly explaining the present invention, and the embodiments of the present invention and the described terms are the spirit of the following claims And it is obvious that various changes and changes can be made without departing from the scope. Such modified embodiments should not be individually understood from the spirit and scope of the present invention, but should be said to fall within the scope of the claims of the present invention.

Claims (17)

As a gaze stabilization control device of a gimbal for receiving a control signal and moving a gaze toward an object,
a following control unit for outputting a first synthesis control signal for compensating for friction within the gimbal according to a command input, a feedback input of the angular velocity of the gimbal, and a gaze error; and
and a stabilization control unit for outputting a second synthesis control signal for compensating for an effect of disturbance applied to the gimbal by using a control signal and the first synthesis control signal. The method according to claim 1,
The following control unit,
a variable estimator for estimating the friction variable of the gimbal from the feedback input of the gimbal; and
and a friction observer for calculating the friction torque of the gimbal using the command input and the estimated friction variable, and outputting the first combined control signal for compensating for the calculated friction torque. The method according to claim 1,
The tracking control unit is a gaze stabilization control device capable of estimating the friction variable of the gimbal using an adaptive control method for sinusoidal driving based on a sliding mode. 4. The method of claim 2 or 3,
The tracking control unit may output an updated first composite control signal according to the feedback input of the gimbal. The method according to claim 1,
The stabilization control unit,
a disturbance detector for extracting a disturbance signal output from a sensor installed in the gimbal; and
and a synthesizer for outputting a second composite control signal by synthesizing the first composite control signal, the control signal, and the disturbance signal. The method according to claim 1,
The stabilization control unit is a gaze stabilization control device capable of compensating for the effect of the disturbance applied to the gimbal in a PI-LEAD control method. As a gaze stabilization control method of a gimbal for receiving a control signal and moving a gaze toward an object,
outputting a first synthesized control signal for compensating for friction within the gimbal according to a command input, a feedback input of the angular velocity of the gimbal, and a gaze error; and
and outputting a second composite control signal for compensating for an effect of disturbance applied to the gimbal by using a control signal and the first composite control signal. 8. The method of claim 7,
The step of outputting the first composite control signal includes the step of estimating the friction variable of the gimbal using an adaptive control method for sinusoidal driving based on a sliding mode. 8. The method of claim 7,
The process of outputting the first composite control signal comprises:
The process of defining a new error (r) by combining the position error and the velocity error for the desired position of the gimbal from the command input as an error equation;
defining a state equation for estimating the internal state variable of the gimbal from the error equation; and
The process of deriving a first composite control signal including the friction variable of the gimbal from the state equation; 10. The method of claim 9,
The error equation is a gaze stabilization control method defined as in Equation 1 below.
[Equation 1]

(From here

, θ is the actual target position, θ _d is the target position, w is the actual angular velocity, and w _d is the target angular velocity. ) 11. The method of claim 10,
The state equation is a gaze stabilization control method defined as in Equation 2 below.
[Equation 2]

(where z ^{^'} is the estimated value of the rate of change dz/dt of the internal state variable, z^ is the estimated value of the internal state variable z, w is the angular velocity of the gimbal, μ ₀ is the design variable, a positive constant value, and sgn is the sign function ( sgn (positive number) = + sign return, sgn (negative number) = - sign return, g(w) denotes a function of angular velocity representing the Strivek effect.) 12. The method of claim 11,
The process of deriving the first composite control signal is a gaze stabilization control method performed by Equation 3 below.
[Equation 3]

(Where u is the first synthesis control signal, k _d and λ are design variables, and when e=θ-θ _d is the position error with respect to the target line of sight (θ _d ), r is the combination of the position error and the speed error. one new error, σ ^{^0} , σ ^{^1} , σ ^{^2} is the estimated friction variable, z^ is the estimate of the gimbal's internal state variable, z^' is the estimate of the rate of change of the gimbal's internal state variable dz/dt, e' is the error between the target angular velocity and the actual angular velocity, θ _d ^'' is the estimate of the target position of the gimbal, and J is the moment of inertia of the system). 13. The method of claim 12,
The process of deriving the first composite control signal includes calculating a friction variable that converges the new error r to 0 using Equations 4 to 6 below.
[Equation 4]

[Equation 5]

[Equation 6]

(Here, γ ₀ , γ ₁ , and γ ₂ are positive constant values as design variables) 14. The method of claim 13,
and updating each of the estimated friction variables σ ^{^ 0} , σ ^{^ 1} , and σ ^{^ 2} according to the feedback input of the gimbal. 8. The method of claim 7,
The process of outputting the second synthesis control signal comprises:
extracting a disturbance signal from a signal output from a sensor installed in the gimbal; and
A gaze stabilization control method comprising: a process of deriving a second synthesized control signal by synthesizing a control signal, a first synthesized control signal, and a disturbance signal. 16. The method of claim 15,
The outputting of the second composite control signal is a gaze stabilization control method using a PI-LEAD control method. 17. A recording medium storing a computer program for performing the gaze stabilization control method according to any one of claims 7 to 16.

Images