Controllability and observability of discretized satellite magnetic attitude control system

Sihui Liu; Qingdao Huang; Sihui Liu; Qingdao Huang

doi:10.3934/math.2023398

AIMS Mathematics

2023, Volume 8, Issue 4: 7899-7916. doi: 10.3934/math.2023398

Previous Article Next Article

Research article

Controllability and observability of discretized satellite magnetic attitude control system

Sihui Liu ,
Qingdao Huang ^,

School of Mathematics, Jilin University, Changchun, 130000, China

Received: 20 October 2022 Revised: 02 January 2023 Accepted: 12 January 2023 Published: 31 January 2023
MSC : 93B05, 34A25

In this paper, two different discrete schemes of the second-order linear time-varying system represented by the linearized satellite magnetic attitude control motion equation are obtained by Euler method. Then, the controllability and observability conditions of a new discrete second-order linear time-varying system are proposed and the validity of these conditions is further verified by some numerical examples. Next, the theoretical results are applied to investigate the controllability and observability of the discretized satellite magnetic control system. Different periods $\tau$ are chosen to investigate the effect on the controllability and observability of the resulting discrete system. The corresponding conclusions are obtained.

Keywords:

linear time-varying system,
controllability,
observability,
discrete second-order system,
satellite attitude control motion equation

Citation: Sihui Liu, Qingdao Huang. Controllability and observability of discretized satellite magnetic attitude control system[J]. AIMS Mathematics, 2023, 8(4): 7899-7916. doi: 10.3934/math.2023398

Related Papers:

[1]	Jingjing Yang, Jianqiu Lu . Stabilization in distribution of hybrid stochastic differential delay equations with Lévy noise by discrete-time state feedback controls. AIMS Mathematics, 2025, 10(2): 3457-3483. doi: 10.3934/math.2025160
[2]	Xiao Yu, Yan Hua, Yanrong Lu . Observer-based robust preview tracking control for a class of continuous-time Lipschitz nonlinear systems. AIMS Mathematics, 2024, 9(10): 26741-26764. doi: 10.3934/math.20241301
[3]	Hany Bauomy . Control and optimization mechanism of an electromagnetic transducer model with nonlinear magnetic coupling. AIMS Mathematics, 2025, 10(2): 2891-2929. doi: 10.3934/math.2025135
[4]	Jenjira Thipcha, Presarin Tangsiridamrong, Thongchai Botmart, Boonyachat Meesuptong, M. Syed Ali, Pantiwa Srisilp, Kanit Mukdasai . Robust stability and passivity analysis for discrete-time neural networks with mixed time-varying delays via a new summation inequality. AIMS Mathematics, 2023, 8(2): 4973-5006. doi: 10.3934/math.2023249
[5]	Limei Liu, Xitong Zhong . Analysis and anti-control of period-doubling bifurcation for the one-dimensional discrete system with three parameters and a square term. AIMS Mathematics, 2025, 10(2): 3227-3250. doi: 10.3934/math.2025150
[6]	Abdul Qadeer Khan, Zarqa Saleem, Tarek Fawzi Ibrahim, Khalid Osman, Fatima Mushyih Alshehri, Mohamed Abd El-Moneam . Bifurcation and chaos in a discrete activator-inhibitor system. AIMS Mathematics, 2023, 8(2): 4551-4574. doi: 10.3934/math.2023225
[7]	Zhiqiang Chen, Alexander Yurievich Krasnov . Disturbance observer based fixed time sliding mode control for a class of uncertain second-order nonlinear systems. AIMS Mathematics, 2025, 10(3): 6745-6763. doi: 10.3934/math.2025309
[8]	Sheng-Ran Jia, Wen-Juan Lin . Adaptive event-triggered reachable set control for Markov jump cyber-physical systems with time-varying delays. AIMS Mathematics, 2024, 9(9): 25127-25144. doi: 10.3934/math.20241225
[9]	Hyung Tae Choi, Jung Hoon Kim . An $L_{\infty}$ performance control for time-delay systems with time-varying delays: delay-independent approach via ellipsoidal $\mathcal{D}$ -invariance. AIMS Mathematics, 2024, 9(11): 30384-30405. doi: 10.3934/math.20241466
[10]	Nasser. A. Saeed, Amal Ashour, Lei Hou, Jan Awrejcewicz, Faisal Z. Duraihem . Time-delayed control of a nonlinear self-excited structure driven by simultaneous primary and 1:1 internal resonance: analytical and numerical investigation. AIMS Mathematics, 2024, 9(10): 27627-27663. doi: 10.3934/math.20241342

Abstract

Table 1. Nomenclature.

$b$	earth's magnetic induction field	$b_0$	earth's magnetic induction field in coordinate system
$e_r$	unit vector of orbital coordinate system	$J_i$	main central inertia moments, $i=1,2,3$
$M_c$	magnetic control moment	$m$	intrinsic magnetic moment
$r$	orbital radius	$\omega$	absolute angular velocity
$\omega_0$	orbital angular velocity	$\theta_i$	attitude Euler angles, $i=1,2,3$
$\mu_E$	earth's magnetic field constant	$\beta$	inclination angle of orbital plane to the equatorial plane
$\Theta$	direction cosine matrix

| Show Table

DownLoad: CSV

1. Introduction

Satellite attitude control system, which is an important guarantee system for satellite normal operation, affects the working performance and on-orbit lifetimes directly. Because of low cost and high reliability, magnetic control technology has been widely applied in the early stages of satellite development ^[1,2,3]. White, Shigemoyo and Bourquin firstly provided the concept of satellite attitude control, making it possible to achieve orientation of satellites with magnetictorquers. Soon, the first satellite magnetic attitude control method was proposed by Ergin and Wheeler ^[4], and they gave some advantages of magnetic control. Later, Renard ^[5] presented method based on averaged models to study the issue of attitude control with magnetic moments. Recently, wide application of periodic control system brings the research and development of magnetic control theory to a new height ^[6,7,8]. In order to predict the trajectory of a satellite with magnetic moments, numerous efforts have been made to investigate the nonlinear system's stability and controllability, which is based on the following Euler dynamic equation ^[9,10]

$\begin{eqnarray} J\frac{d\omega}{dt}+\omega\times{J\omega} = 3\omega_0^2(e_r\times{Je_r})+M_c, \end{eqnarray}$

(1.1)

where $J = {\rm{diag(J_1, J_2, J_3) }}$ , $\omega = [\omega_1, \omega_2, \omega_3]^T$ , $\omega_0 = [\omega^0_1, \omega^0_2, \omega^0_3]^T$ , the unit vector $e_r = [0, 0, −1]$ , and $\times$ is the familiar operation of cross product. The control moment $M_c$ is

$\begin{eqnarray} M_c = m(t)\times{b(t)}, \end{eqnarray}$

where $m(t) = [m_1, m_2, m_3]^T$ and $b(t) = \Theta b_0(t)$ , $b_0(t)$ can be approximated by the direct magnetic dipole as ^[11],

$\begin{eqnarray} b_0(t) = \frac{\mu_E}{r^3}\left[ \begin{array}{ccc} \cos\omega_0{t}\sin{\beta} \\ -\cos{\beta}\\ 2\sin\omega_0{t}\sin{\beta} \end{array} \right]. \end{eqnarray}$

Many scholars focused on controllability of satellite magnetic control system based on control theories for nonlinear and linear time-varying continuous systems ^{[12,13,14,15]}. However, exact solutions and fundamental solution matrix of these time-varying continuous systems are hard to obtain, although controllability conditions of systems are satisfied. Therefore, the trajectory of satellites can not be predicted correctly. Therefore, numerous studies have been done to deal with the nonlinear second-order time-varying system (1.1) by linearization method ^[13,15]. With the development of computer technology, quite a number of research works regarding discrete-time systems have been reported in the literature ^[16,17,18] and the controllability and observability of discrete linear systems have attracted a lot of interest ^{[19,20,21,22,23]}. Witczak etc. provided a necessary and sufficient condition for the observability of first-order discrete time-varying linear systems. Mahmudov proposed the controllability and observability conditions of second-order discrete linear time-varying systems in a matrix form. The controllability and observability conditions for the problem of discrete satellite magnetic attitude control have not been presented to the best of our knowledge. Usually, the difference methods based on Taylor series are widely used for approximation discretization of the continuous-time systems ^[23]. A brief review of the possible approach to discretize linear and nonlinear time-varying systems so far has been presented in ^[24]. Here, we investigate controllability and observability property of the linearized form of equation (1.1) by transforming it into a discrete time-varying system with second derivation by the forward and backward Euler method. Then, the controllability and observability conditions of a new discrete second-order linear time-varying system are proposed, which are applied to investigate the controllability and observability of the discretized satellite magnetic control system. Different periods $\tau$ are chosen to investigate the effect on controllability and observability of the resulting discrete system.

The rest of this paper is structured as follows. In Section 2, two different discrete schemes of the second-order linear time-varying system represented by the linearized satellite magnetic attitude control motion equation are obtained by Euler method. The linearized satellite magnetic attitude control system is changed into a discrete second-order time-varying system. In Section 3, the controllability conditions of a new discrete second-order linear time-varying system are proposed, which are applied to investigate the controllability of the discretized satellite magnetic control system. Section 4 investigates the observability of the discrete second-order satellite magnetic control system based on corresponding observability conditions. We give concluding remarks in the final section.

2. Transformation of satellite attitude control system

In this section, the nonlinear second-order time-varying system represented by satellite attitude magnetic control motion equation (1.1) is linearized. We obtain two different discrete schemes of the second-order linear time-varying system by Euler method and transform the linearized satellite magnetic attitude control motion equation into a discrete second-order time-varying linear system.

2.1. Linearized equation of satellite magnetic control

We assume the mass center of the satellite moving in earth's gravitational field and in a circular orbit. To satellite magnetic attitude control motion equation, the coordinate systems are described as follows:

(1) orbital system $(X, Y, Z).$ The origin is satellite mass center. The $Z$ -axis points in the direction of the radius vector; the $Y$ -axis is normal to the satellite orbit plane; and the $X$ -axis forms the right-hand trial.

(2) satellite body frame $(x, y, z)$ . The axes are assumed to coincide with the body's principle inertia axes and their origin are still at the center of satellite mass.

The attitude of system $(x, y, z)$ relative to the orbital system $(X, Y, Z)$ is given by Euler angles $\theta_1, \; \theta_2, \; \theta_3.$ Then the components of $\omega$ have form ^[13]

$\begin{eqnarray} \omega = \left[ \begin{array}{ccc} \omega_1 \\ \omega_2\\ \omega_3 \end{array} \right] = C_{\theta_3\theta_2}\left[ \begin{array}{ccc} \dot{\theta}_1 \\ \dot{\theta}_2\\ \dot{\theta}_3 \end{array} \right]+\Theta\left[ \begin{array}{ccc} \omega_1^0 \\ \omega_2^0\\ \omega_3^0 \end{array} \right], \end{eqnarray}$

(2.1)

where

$\begin{eqnarray} C_{\theta_3\theta_2} = \left[ \begin{array}{ccc} \cos\theta_2\cos\theta_3&\sin\theta_3&0\\ \cos\theta_2\sin\theta_3&\cos\theta_3&0\\ \sin\theta_2&0&1 \end{array} \right], \end{eqnarray}$

(2.2)

and the direction cosine matrix $\Theta = [\Theta_{ij}]$ is represented using a 1-2-3 Euler angle rotation sequence as follows

$\begin{eqnarray} \left\{ \begin{array}{lllllllll} &\Theta_{11} = \cos\theta_2\cos\theta_3, \\ &\Theta_{12} = \cos\theta_1\sin\theta_3+\sin\theta_1\sin\theta_2\cos\theta_3, \\ &\Theta_{13} = \sin\theta_1\sin\theta_3-\cos\theta_1\sin\theta_2\cos\theta_3, \\ &\Theta_{21} = -\cos\theta_2\sin\theta_3, \\ &\Theta_{22} = \cos\theta_1\cos\theta_3-\sin\theta_1\sin\theta_2\sin\theta_3, \\ &\Theta_{23} = \sin\theta_1\cos\theta_3+\cos\theta_1\sin\theta_2\sin\theta_3, \\ &\Theta_{31} = \sin\theta_2, \\ &\Theta_{32} = \sin\theta_1\cos\theta_2, \\ &\Theta_{33} = \cos\theta_1\cos\theta_2. \end{array} \right. \end{eqnarray}$

(2.3)

In ^[15], Morozov and Kalenova gave the following linearized special second-order time-varying system of equation (1.1)

$\begin{eqnarray} \ddot{x}(t)+D\dot{x}(t)+Kx(t) = B(t)u(t), \end{eqnarray}$

(2.4)

where

$\begin{eqnarray} D = \left[ \begin{array}{ccc} 0&0&\frac{J_{1}+J_{32}}{J_{1}}\omega_0 \\ 0&0&0\\ \frac{J_{21}-{J_3}}{J_3}\omega_0&0&0 \end{array} \right] = \left[ \begin{array}{ccc} 0&0&-d_1\omega_0 \\ 0&0&0\\ d_3\omega_0&0&0 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} K = \left[ \begin{array}{ccc} \frac{-4\omega_0^2J_{32}}{J_{1}}&0&0 \\ 0&\frac{3\omega_0^2J_{13}}{J_2}&0\\ 0&0&\frac{\omega_0^2J_{21}}{J_3} \end{array} \right] = \left[ \begin{array}{ccc} -k_1\omega_0^2&0 \\ 0&-k_2\omega_0^2&0\\ 0&0&-k_3\omega_0^2 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} B(t) = \omega_0^2\epsilon\left[ \begin{array}{ccc} 0&2b_1\sin\omega_0{t}&b_4 \\ -2b_2\sin\omega_0{t}&0&b_2\cos\omega_0{t}\\ -b_5&-b_3\cos\omega_0{t}&0 \end{array} \right], \end{eqnarray}$

and

$\begin{eqnarray} \begin{array}{lllll} &x(t) = [\theta_1, \; \theta_2, \; \theta_3]^T{is\; the\; state\; vector}, \\ &\epsilon = \frac{\mu_E}{\omega^2_0{r^3}}, \; \; \; d = J_2-J_1+J_3, \; \; \; \; d_i = \frac{d}{J_i}\; (i = 1, 3), \; \; \; J_{ij} = J_i-J_j\; (i, j = 1, 2, 3), \\ &k_1 = \frac{-4J_{32}}{J_{1}}, \; \; \; \; k_2 = \frac{3J_{31}}{J_2}, \; \; \; \; k_3 = \frac{J_{12}}{J_3}, \\ &b_j = \frac{\sin{\beta}}{J_j}\; (j = 1, 2, 3), \; \; \; \; b_4 = \frac{\cos{\beta}}{J_1}, \; \; \; \; b_5 = \frac{\cos{\beta}}{J_3}\\ &{u(t) = m(t) = [m_1, \; m_2, \; m_3]^T\; is\; control\; vector, \; which\; allows\; the\; satellite\; attitude\; position\; to\; be\; stabilized}. \end{array} \end{eqnarray}$

The complete linearized derivation of equation (1.1) can be found in ^[13].

With certain assumptions, the equation of the measurement yielded by the magnetometer is represented as ^[25]

$\begin{eqnarray} y(t) = C(t)x(t) \end{eqnarray}$

(2.5)

where

$\begin{eqnarray} C(t) = \left[ \begin{array}{lll} \alpha_2&0&-\alpha_1\sin\omega_0{t}-\alpha_3\cos\omega_0{t}\\ 0&-\alpha_2&\alpha_1\cos\omega_0{t}-\alpha_3\sin\omega_0{t} \end{array} \right] \end{eqnarray}$

and $\alpha_j\; (j = 1, 2, 3)$ are constant quantities determining the position of the orbit in space, $x(t)\in R^3, \; y(t)\in R^2$ are state vector and output vector respectively.

Then we have the following linear satellite magnetic attitude control system with measurement, corresponding coefficient matrices are in accordance with matrices in equation (2.4) and (2.5)

$\begin{eqnarray} \left\{ \begin{array}{ll} &\ddot{x}(t)+D\dot{x}(t)+Kx(t) = B(t)u(t), \\ &y(t) = C(t)x(t). \end{array} \right. \end{eqnarray}$

(2.6)

2.2. Discretization of the linearized satellite control system

Firstly, letting $t = k\tau\; (k = 0, 1, 2...),$ we realize an approximate form of $\dot{x}(t)$ and $\ddot{x}(t)$ using Taylor forward expansion as

$\begin{equation} \dot{x}(t)\approx\frac{x((k+1)\tau)-x(k\tau)}{\tau} \end{equation}$

(2.7)

and

$\begin{equation} \ddot{x}(t)\approx\frac{x((k+1)\tau)-2x(k\tau)+x((k-1)\tau)}{\tau^2}. \end{equation}$

(2.8)

Substituting equation (2.7) and (2.8) into system (2.6), we have

$\begin{equation} \left\{ \begin{array}{ll} &\frac{x((k+1)\tau)-2x(k\tau)+x((k-1)\tau)}{\tau^2}+D\frac{x((k+1)\tau)-x(k\tau)}{\tau}+Kx(k\tau) = B(k\tau)u(k\tau), \\ &y(k\tau) = C(k\tau)x(k\tau). \end{array} \right. \end{equation}$

(2.9)

Then we can get

$\begin{equation} \left\{ \begin{array}{ll} &(I_3+\tau D)x_{k+1}+(\tau^2 K-\tau D-2I_3)x_k+x_{k-1} = \tau^2B_ku_k, \\ &y_k = C_kx_k, \end{array} \right. \end{equation}$

(2.10)

where $I_3$ denotes the identity matrix of $3$ dimension and

$\begin{eqnarray} x_k = x(k\tau), \; \; u_k = u(k\tau), \; \; y_k = y(k\tau), \; \; C_k = C(k\tau), \; \; B_k = B(k\tau). \end{eqnarray}$

(2.11)

Noting that

$\begin{eqnarray} I_3+\tau D = \left[ \begin{array}{ccc} 1&0&-d_1\omega_0\tau \\ 0&1&0\\ d_3\omega_0\tau&0&1 \end{array} \right] \end{eqnarray}$

(2.12)

and

$\begin{eqnarray} det(I_3+\tau D) = 1+d_1d_3\omega^2_0\tau^2 > 0, \end{eqnarray}$

(2.13)

which means the matrix $\; I_3+\tau D\;$ is invertible. Therefore, we can rewrite the system (2.10) as the following discrete system

$\begin{equation} \left\{ \begin{array}{ll} &x_{k+1} = A_0x_{k-1}+A_1x_k+\widetilde{B}_ku_k, \\ &y_k = C_kx_k, \end{array} \right. \end{equation}$

(2.14)

where

$\begin{eqnarray} A_0 = -(I_3+\tau D)^{-1} = \left[ \begin{array}{ccc} -\frac{1}{d_1d_3\omega^2_0\tau^2+1}&0&-\frac{d_1\omega_0\tau}{d_1d_3\omega^2_0\tau^2+1} \\ 0&-1&0\\ \frac{d_3\omega_0\tau}{d_1d_3\omega^2_0\tau^2+1}&0&-\frac{1}{d_1d_3\omega^2_0\tau^2+1} \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} A_1 = (I_3+\tau D)^{-1}(\tau D+2I_3-\tau^2 K) = \left[ \begin{array}{ccc} 1+\frac{k_1\omega^2_0\tau^2+1}{d_1d_3\omega^2_0\tau^2+1}&0&\frac{d_1\tau\omega_0(k_3\omega^2_0\tau^2+1)}{d_1d_3\omega^2_0\tau^2+1} \\ 0&k_2\omega_0^2 \tau^2+2&0\\ \frac{-d_3\omega_0\tau(k_1\omega^2_0\tau^2+1)}{d_1d_3\omega^2_0\tau^2+1}&0&1+\frac{k_3\omega^2_0\tau^2+1}{d_1d_3\omega^2_0\tau^2+1} \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} \widetilde{B}_k = (I_3+\tau D)^{-1}\tau^2B_k = \tau^2\omega_0^2\epsilon\left[ \begin{array}{ccc} -\frac{b_5 d_1\omega_0\tau}{d_1d_3\omega^2_0\tau^2+1}&\frac{2b_1\sin\omega_0{k\tau}-b_3 d_1\omega_0\tau \cos\omega_0{k\tau}}{d_1d_3\omega^2_0\tau^2+1}&\frac{b_4}{d_1d_3\omega^2_0\tau^2+1} \\ -2b_2\sin\omega_0{k\tau}&0&b_2\cos\omega_0{k\tau}\\ -\frac{b_5}{d_1d_3\omega^2_0\tau^2+1}&\frac{-(b_3\cos\omega_0{k\tau}+2b_1d_3\omega_0\tau\sin\omega_0k\tau)}{d_1d_3\omega^2_0\tau^2+1}&\frac{-b_4d_3\omega_0\tau}{d_1d_3\omega^2_0\tau^2+1} \end{array} \right]. \end{eqnarray}$

Next, by analogy, the system (2.6) can also be discretized with backward Euler method. We can obtain an approximate form of $\dot{x}(t)$ and $\ddot{x}(t)$ using Taylor backward expansion as

$\begin{equation} \dot{x}(t)\approx\frac{x(k\tau)-x((k-1)\tau)}{\tau} \end{equation}$

(2.15)

and

$\begin{equation} \ddot{x}(t)\approx\frac{x((k+1)\tau)-2x(k\tau)+x((k-1)\tau)}{\tau^2}. \end{equation}$

(2.16)

Substituting equation (2.15) and (2.16) into system (2.6), letting $t = k\tau\; (k = 0, 1, 2...),$

$\begin{equation} \left\{ \begin{array}{ll} &\frac{x((k+1)\tau)-2x(k\tau)+x((k-1)\tau)}{\tau^2}+D\frac{x((k\tau)-x((k-1)\tau)}{\tau}+Kx(k\tau) = B(k\tau)u(k\tau), \\ &y(k\tau) = C(k\tau)x(k\tau), \end{array} \right. \end{equation}$

(2.17)

Based on equation (2.11) and (2.17) we get

$\begin{equation} \left\{ \begin{array}{ll} &x_{k+1}+(K\tau^2+D\tau-2I_3)x_k+(I_3-D\tau)x_{k-1} = \tau^2B_ku_k, \\ &y_k = C_kx_k. \end{array} \right. \end{equation}$

(2.18)

Therefore, we can rewrite the system (2.17) as the following discrete system with backward Euler method,

$\begin{equation} \left\{ \begin{array}{ll} &x_{k+1} = \widetilde{A}_0x_{k-1}+\widetilde{A}_1x_k+\widetilde{\widetilde{B}}_ku_k, \\ &y_k = C_kx_k, \end{array} \right. \end{equation}$

(2.19)

where

$\begin{eqnarray} \widetilde{A}_0 = \tau D-I_3 = \left[ \begin{array}{ccc} -1&0&-d_1\omega_0\tau \\ 0&-1&0\\ d_3\omega_0\tau&0&-1 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} \widetilde{A}_1 = 2I_3-\tau D-\tau^2 K = \left[ \begin{array}{ccc} k_1\omega^2_0\tau^2+2&0&d_1\omega_0\tau \\ 0&k_2\omega_0^2 \tau^2+2&0\\ -d_3\omega_0\tau&0&k_3\omega^2_0\tau^2+2 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} \widetilde{\widetilde{B}}_k = \tau^2\omega_0^2\epsilon\left[ \begin{array}{ccc} 0&2b_1\sin\omega_0{k\tau}&b_4 \\ -2b_2\sin\omega_0{k\tau}&0&b_2\cos\omega_0{k\tau}\\ -b_5&-b_3\cos\omega_0{k\tau}&0 \end{array} \right], \; k = 0, 1, 2...\; . \end{eqnarray}$

In general, when the sampling period $\tau$ is about one tenth of the minimum time constant of the system, the approximation is satisfactory enough. In fact, the error between the exact solution and the numerical solution will be lager if the period is too big, which makes the mathematical precision lower. On the contrary, if the period is too small, then step size increases and the computation is huge. Then, based on the similar form of discrete system (2.14) and (2.19), we can rewrite them as the following general discrete system

$\begin{equation} \left\{ \begin{array}{ll} &x_{k+1} = \hat{A}_0x_{k-1}+\hat{A}_1x_k+\hat{B}_ku_k, \\ &y_k = C_kx_k, \end{array} \right. \end{equation}$

(2.20)

where $x_k\in R^n, \; y_k\in R^r, \; u_k\in R^m\; (m\leq n)$ are the state vector, the output vector and the control vector respectively. $\hat{A}_0, \; \hat{A}_1\in R^{n\times n}$ , $\hat{B}_k\in R^{n\times m}, \; C_k\in R^{r\times n}$ are coefficient matrices.

3. Controllability of the discrete system

3.1. Controllability analysis

According to the definition of controllability of discrete linear time-varying systems in ^[26], we present the definition of controllability and uncontrollability of the second-order discrete time-varying linear system (2.20).

Definition 1. The second-order discrete time-varying linear system (2.20) is said to be controllable to final state $x_n = x_f$ in finite $n$ steps if it exists an input sequence $U = \{u_0, u_1, ..., u_{n-1}\}$ which brings the initial state $(x_{-1}, x_0)$ to a final state $x_n = x_f$ in the finite discrete time interval $[0, n].$ Otherwise the system (2.20) is uncontrollable.

Theorem 1. The linear discrete-time varying system (2.20) is controllable if and only if ${\rm{rank}}\; \mathbb{C} = {\rm{n}}$ , and the controllability matrix $\mathbb{C}$ is defined by

$\begin{equation} {\mathbb{C} = [M^{(n)}_{n-1}\hat{B}_{n-1}, \; M^{(n)}_{n-2}\hat{B}_{n-2}, ..., M^{(n)}_{1}\hat{B}_{1}, M^{(n)}_{0}\hat{B}_0], } \end{equation}$

(3.1)

where

$\begin{eqnarray} \begin{array}{ll} &M^{(n)}_{i-2} = M^{(n)}_{i-1}\hat{A}_1+M^{(n)}_{i}\hat{A}_0, \; \; i = 2, ..., n-1, \\ &M^{(n)}_{n-2} = \hat{A}_1, \; \; \; M^{(n)}_{n-1} = I_n. \end{array} \end{eqnarray}$

(3.2)

Proof Based on the form of the state vector in the system (2.20), we derive the following set of equations

$\begin{eqnarray} \begin{array}{ll} &{\; x_1 = \hat{A}_0x_{-1}+\hat{A}_1x_0+\hat{B}_0u_0, }\\ &\begin{array}{ll} x_2& = \hat{A}_0x_0+\hat{A}_1x_1+\hat{B}_1u_1 = \hat{A}_0x_1+\hat{A}_1(\hat{A}_0x_{-1}+\hat{A}_1x_0+\hat{B}_0u_0)+\hat{B}_1u_1\\ & = \hat{A}_1\hat{A}_0x_{-1}+(\hat{A}_0+\hat{A}^2_1)x_0+\hat{A}_1\hat{B}_0u_0+\hat{B}_1u_1. \end{array} \end{array} \end{eqnarray}$

(3.3)

Thus, by iteration, it is finally not difficult to find that in equations (3.3) the $k-th$ instant state vector $x_k$ starting from initial time $k = 0$ and $k = 1$ is

$\begin{equation} x_k = {Q(k)x_{-1}+P(k)x_0+\sum\limits_{i = 0}^{k-1}M_i^{(k)}\hat{B}_iu_i}, \end{equation}$

(3.4)

where $Q(k), \; P(k), \; M_i^{(k)}$ are polynomial functions consisted of matrices $\hat{A}_0, \; \hat{A}_1$ , and they satisfy the following important iteration equations

$\begin{eqnarray} \begin{array}{lll} &Q(k) = \hat{A}_0Q(k-2)+\hat{A}_1Q(k-1), \; \; Q(1) = O_{n\times n}, \; \; Q(0) = I_n, \\ &P(k) = \hat{A}_0P(k-2)+\hat{A}_1P(k-1), \; \; P(0) = O_{n\times n}, \; \; P(1) = I_n, \; \; k = 2, 3, ...\; , \\ &M_{i-2}^{(k)} = M_{i-1}^{(k)}\hat{A}_1+M_i^{(k)}\hat{A}_0, \; \; i = 2, ..., k-1, \; \; M_{k-2}^{(k)} = \hat{A}_1, \; \; \; M_{k-1}^{(k)} = I_n. \end{array} \end{eqnarray}$

(3.5)

In a general case, the state vector at final time $n$ can be written as

$\begin{equation} x_{n} = {Q(n)x_{-1}+P(n)x_0+\sum\limits_{i = 0}^{n-1}M_i^{(n)}\hat{B}_iu_i}. \end{equation}$

(3.6)

Therefore

$\begin{equation} \begin{array}{lll} &x_{n}-Q(n)x_{-1}-P(n)x_0 = \sum\limits_{i = 0}^{n-1}M_i^{(n)}\hat{B}_iu_i\\ & = {M_0^{(n)}\hat{B}_0u_0+M_1^{(n)}\hat{B}_1u_1+...+M_{n-1}^{(n)}\hat{B}_{n-1}u_{n-1}}\\ & = [M_{n-1}^{(n)}\hat{B}_{n-1}, \; M_{n-2}^{(n)}\hat{B}_{n-2}, ..., M_1^{(n)}\hat{B}_{1}, M_0^{(n)}\hat{B}_0]{\left[ \begin{array}{lllll} u_{n-1}\\ u_{n-2}\\ ...\\ u_1\\ u_0 \end{array} \right].} \end{array} \end{equation}$

(3.7)

If we denote

$\begin{eqnarray} {\mathbb{C} = [M_{n-1}^{(n)}\hat{B}_{n-1}, \; M_{n-2}^{(n)}\hat{B}_{n-2}, ..., M_1^{(n)}\hat{B}_{1}, M_0^{(n)}\hat{B}_0]} \end{eqnarray}$

as the controllability matrix in the sequel, then the equation (3.7) determines the input sequence which transfers the initial state $(x_{-1}, x_0)$ to the desired state $x_f = x_n$ in $n$ steps. Thus, these equations will have a solution for any given vector $x_f$ if and only if the matrix has full rank, i.e. ${\rm{rank}}\; \mathbb{C} = {\rm{n}},$ the discrete system (2.20) is controllable.

In order to further verify the correctness of the theoretical results, some numerical examples are given.

Example.1 Let us investigate the controllability of the following system ^[20]:

$\begin{eqnarray} \ddot{x}(t) = \left[ \begin{array}{ccc} 1&0&-1\\ 2&3&1\\ 4&-5&2 \end{array} \right]x(t) +\left[ \begin{array}{lll} 1\\ 0\\ 3 \end{array} \right]u(t). \end{eqnarray}$

(3.8)

Firstly, it is noticed that there is no $\dot{x}(t)$ in system (3.8), which means the discrete forms with forward and backward method of system (3.8) are equivalent, that is

$\begin{eqnarray} x_{k+1} = \hat{A}_0x_{k-1}+\hat{A}_1x_k+\hat{B}u_k \end{eqnarray}$

(3.9)

where

$\begin{eqnarray} \hat{A}_1 = 2I_3+\tau^2A, \; \; \; \; \hat{A}_0 = -I_3, \; \; \; \; \hat{B} = \tau^2B, \; \; \; \; A = \left[ \begin{array}{ccc} 1&0&-1\\ 2&3&1\\ 4&-5&2 \end{array} \right], \; \; B = \left[ \begin{array}{lll} 1\\ 0\\ 3 \end{array} \right]. \end{eqnarray}$

Letting $\tau = 0.1s,$ according to Theorem $1$ , we have

$\begin{eqnarray} n = 3, \; \; M_2^{(3)} = I_3, \; \; M_1^{(3)} = \hat{A}_1 = \left[ \begin{array}{ccc} 2.01&0&-0.01\\ 0.02&2.03&0.01\\ 0.04&-0.05&2.02 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} M_0^{(3)} = M_1^{(3)}\hat{A}_1+M_2^{(3)}\hat{A}_0 = \left[ \begin{array}{ccc} 3.0397&0.0005&-0.0403\\ 0.0812&3.1204&0.0403\\ 0.1602&-0.2025&3.0795 \end{array} \right]. \end{eqnarray}$

Then

$\begin{eqnarray} \mathbb{C} = [M^{(3)}_{2}\hat{B}, \; M^{(3)}_{1}\hat{B}, M^{(3)}_{0}\hat{B}] = \tau^2[M^{(3)}_{2}B, \; M^{(3)}_{1}B, M^{(3)}_{0}B] = \tau^2[B, \; \hat{A}_1B, \; (\hat{A}_1^2+\hat{A}_0)B], \end{eqnarray}$

(3.10)

here,

$\begin{eqnarray} \hat{A}_1B = \left[ \begin{array}{lll} 1.98\\ 0.05\\ 6.1 \end{array} \right], \; \; \; (\hat{A}_1^2+\hat{A}_0)B = \left[ \begin{array}{lll} 2.9188\\ 0.2021\\ 9.3987 \end{array} \right], \end{eqnarray}$

and

$\begin{eqnarray} det\; \mathbb{C} = -0.000221\neq 0, \; \; \; rank\; \mathbb{C} = 3 = n. \end{eqnarray}$

(3.11)

Therefore, discrete system (3.9) is controllable.

Example.2 The equations of controlling the motion of a spacecraft between the earth and the moon have the form ^[25]

$\begin{eqnarray} \ddot{x}(t)+2D\dot{x}(t)+Kx(t) = B(t)u(t) \end{eqnarray}$

(3.12)

Here,

$\begin{eqnarray} D = \left[ \begin{array}{cc} 0&-1\\ 1&0 \end{array} \right], \; \; K = \left[ \begin{array}{cc} -\alpha_1&0\\ 0&-\alpha_2 \end{array} \right], \; \; B(t) = \left[ \begin{array}{cc} -\cos t\\ \sin t \end{array} \right], \; \; \alpha_1, \alpha_2 = const. \end{eqnarray}$

For forward Euler method, the discrete form of system is as follows:

$\begin{eqnarray} (I+2\tau D)x_{k+1}+(\tau^2K-2I-2\tau D)x_k+x_{k-1} = \tau^2 B_ku_k. \end{eqnarray}$

(3.13)

Then, choosing $\tau = 0.1s,$ we have

$\begin{eqnarray} I+2\tau D = \left[ \begin{array}{cc} 1&-2\tau\\ 2\tau&1 \end{array} \right] = \left[ \begin{array}{cc} 1&-0.2\\ 0.2&1 \end{array} \right], \; \; det(I+2\tau D) = 1+4\tau^2 = 1.04\neq 0, \end{eqnarray}$

which means the inverse matrix of $I+2\tau D$ exits and

$\begin{eqnarray} (I+2\tau D)^{-1} = \left[ \begin{array}{cc} \frac{1}{1+4\tau^2}&\frac{2\tau}{1+4\tau^2}\\ -\frac{2\tau}{1+4\tau^2}&\frac{1}{1+4\tau^2} \end{array} \right]\approx\left[ \begin{array}{cc} 0.9615&0.1923\\ -0.1923&0.9615 \end{array} \right]. \end{eqnarray}$

The discrete system can be represented as

$\begin{eqnarray} x_{k+1} = \hat{A}_0x_{k-1}+\hat{A}_1x_k+\hat{B}_ku_k \end{eqnarray}$

(3.14)

where

$\begin{eqnarray} \hat{A}_0 = -(I+2\tau D)^{-1}, \; \; \hat{A}_1 = (I+2\tau D)^{-1}(2I+2\tau D-\tau^2K), \; \; \hat{B}_k = \tau^2 (I+2\tau D)^{-1}B(k\tau). \end{eqnarray}$

According to Theorem $1$ , we have

$\begin{eqnarray} n = 2, \; \; M_1^{(2)} = I_2, \; \; M_0^{(2)} = \hat{A}_1. \end{eqnarray}$

Then

$\begin{eqnarray} \mathbb{C} = [M^{(2)}_{1}\hat{B}_{1}, \; \; M^{(2)}_{0}\hat{B}_0] = [\hat{B}_1, \; \hat{A}_1\hat{B}_0], \end{eqnarray}$

(3.15)

here

$\begin{eqnarray} \hat{B}_1\approx\left[ \begin{array}{ll} 0.00937\\ 0.00287 \end{array} \right], \; \; \; \hat{B}_0\approx\left[ \begin{array}{cc} -0.00962\\ 0.00192 \end{array} \right], \; \; \; \hat{A}_1 = \left[ \begin{array}{cc} \frac{2.04+0.01\times \alpha_1}{1.04}&\frac{0.2+0.002\times \alpha_2}{1.04}\\ \frac{-0.2-0.002\times \alpha_1}{1.04}&\frac{2.04+0.01\times \alpha_2}{1.04} \end{array} \right]. \end{eqnarray}$

Because of $\alpha_1 = 1+2b, \; \; \alpha_2 = 1-b, \; \; 0 < b < < 1$ , then

$\begin{eqnarray} det\; \mathbb{C}\approx-0.000112-0.0000021\alpha_2\neq 0, \; \; \; rank\; \mathbb{C} = 2 = n. \end{eqnarray}$

(3.16)

Therefore, discrete system (3.13) is controllable. Similarly, we can prove the controllability of system (3.12) with forward Euler discretized form.

3.2. Numerical calculations of controllability of discrete magnetic attitude control system

To the specific discrete system (2.19) with backward difference, we have

$\begin{eqnarray} M_2^{(3)} = I_3, \end{eqnarray}$

$\begin{eqnarray} M_1^{(3)} = \hat{A}_1 = \widetilde{A}_1 = \left[ \begin{array}{ccc} k_1\omega^2_0\tau^2+2&0&d_1\tau\omega_0 \\ 0&k_2\omega_0^2 \tau^2+2&0\\ -d_3\tau\omega_0&0&k_3\omega^2_0\tau^2+2 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} { \begin{array}{ll} &M_0^{(3)} = M_1^{(3)}\hat{A}_1+M_2^{(3)}\hat{A}_0 = \widetilde{A}_1^2+\widetilde{A}_0 = \\ &\left[ \begin{array}{ccc} (k_1\omega^2_0\tau^2+2)^2-d_1d_3\tau^2\omega_0^2-1&0&d_1\tau\omega_0(k_1\omega^2_0\tau^2+k_3\omega^2_0\tau^2+3) \\ 0&{(k_2\omega_0^2 \tau^2+2)}^2-1&0\\ -d_3\tau\omega_0(k_1\omega^2_0\tau^2+k_3\omega^2_0\tau^2+3)&0&-d_1d_3\tau^2\omega_0^2+(k_3\omega^2_0\tau^2+2)^2-1 \end{array} \right], \end{array}} \end{eqnarray}$

$\begin{eqnarray} \hat{B}_2 = \widetilde{\widetilde{B}}_2 = \tau^2\omega_0^2\epsilon\left[ \begin{array}{ccc} 0&2b_1\sin2\omega_0{\tau}&b_4 \\ -2b_2\sin2\omega_0{\tau}&0&b_2\cos2\omega_0{\tau}\\ -b_5&-b_3\cos2\omega_0{\tau}&0 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} \hat{B}_1 = \widetilde{\widetilde{B}}_1 = \tau^2\omega_0^2\epsilon\left[ \begin{array}{ccc} 0&2b_1\sin\omega_0{\tau}&b_4 \\ -2b_2\sin\omega_0{\tau}&0&b_2\cos\omega_0{\tau}\\ -b_5&-b_3\cos\omega_0{\tau}&0 \end{array} \right]. \end{eqnarray}$

$\begin{eqnarray} { \hat{B}_0 = \widetilde{\widetilde{B}}_0 = \tau^2\omega_0^2\epsilon\left[ \begin{array}{ccc} 0&0&b_4 \\ 0&0&b_2\\ -b_5&-b_3&0 \end{array} \right].} \end{eqnarray}$

$\begin{eqnarray} M_1^{(3)}\hat{B}_1 = \widetilde{A}_1\widetilde{\widetilde{B}}_1 = \tau^2\omega_0^2\epsilon\left[ \begin{array}{ccc} -b_5d_1\omega_0\tau&l_{12}&b_4(k_1\omega^2_0\tau^2+2) \\ -2b_2\sin\omega_0{\tau}(k_2\omega^2_0\tau^2+2)&0&b_2\cos\omega_0{\tau}(k_2\omega^2_0\tau^2+2)\\ -b_5(k_3\omega^2_0\tau^2+2)&l_{32}&-b_4d_3\omega_0\tau \end{array} \right] \end{eqnarray}$

where

$\begin{eqnarray} \left\{ \begin{array}{ll} &l_{12} = 2b_1\sin\omega_0{\tau}(k_1\omega^2_0\tau^2+2)-b_3d_1\omega_0\tau\cos\omega_0\tau, \\ &l_{32} = -2b_1d_3\omega_0\tau\sin\omega_0\tau-b_3\cos\omega_0{\tau}(k_3\omega^2_0\tau^2+2). \end{array} \right. \end{eqnarray}$

It can be shown

$\begin{eqnarray} {\rm{det}}{\rm{(M_2^{(3)}\hat{B}_2) = det(\hat{B}_2) = b_2(b_3b_4-b_1b_5)\sin4}}\omega_0\tau = 0 \end{eqnarray}$

(3.17)

where

$\begin{eqnarray} b_3b_4-b_1b_5 = \frac{\sin{\beta}}{J_3}\frac{\cos{\beta}}{J_1}-\frac{\sin{\beta}}{J_1}\frac{\cos{\beta}}{J_3} = 0. \end{eqnarray}$

(3.18)

Based on the results in equation (3.17), the condition of full rank of $\mathbb{C}$ is not satisfied when the determinant of matrix $\mathbb{C}$ composed of the first three columns. However, if the other three independent columns (such as, 1, 2, 5) are selected, for convenience, we choose $\tau = 0.1$ as the period of the discrete system without loss of generality, it is easy to see that ${\rm{det}}(\mathbb{C})\neq 0$ on the basic of satellite and its orbit parameters in ^[27]. Then

$\begin{equation} { {\rm{rank}}\; \mathbb{C} = {\rm{rank\; [M_2^{(3)}\hat{B}_2, \; \; M_1^{(3)}\hat{B}_1, \; \; M_0^{(3)}\hat{B}_0] = 3.}}} \end{equation}$

(3.19)

According to Theorem $1$ , the system (2.19) by backward Euler method is controllable.

Moreover, we make $\tau = 0.2, \; 0.05$ respectivly to compute the rank of $\mathbb{C}$ and the results show different $\tau$ have no effect on the controllability of system (2.19). In the same way, we can also calculate the rank of $\mathbb{C}$ for the system (2.14) with Matlab and then prove the discretized system (2.14) by forward Euler method is controllable.

4. Observability of the discrete system

4.1. Observability analysis

Based on the definition of observability of discrete linear time-varying systems in ^[28], we can also present the following definition of observability of system (2.20).

Definition 2. The second order discrete linear time-varying system (2.20) is observable if for any unknown initial state $(x_{0}, \; x_1)$ , there exists a finite $k_{\beta}\in N(k_{\beta} > 0)$ such that $(x_{0}, \; x_1)$ can be determined uniquely from the knowledge of output $y_k$ and input $u_k, \; k\in[0, k_{\beta}].$ Otherwise the system is said to be unobservable.

Theorem 2. The second-order linear discrete-time system (2.20) is observable if and only if the observability matrix $\mathbb{S}$ has rank equal to $2n$ and

$\begin{equation} \nonumber \mathbb{S} = \left[ \begin{array}{llllll} C_0&O\\ O&C_1\\ C_2\hat{A}_0&C_2\hat{A}_1\\ C_3\hat{A}_1\hat{A}_0&C_3(\hat{A}_0+\hat{A}_1^2)\\ ...&...\\ C_{2n-1}Q(2n-1)&C_{2n-1}P(2n-1) \end{array} \right] \end{equation}$

where the definitions of $M(k), P(k)$ are same as above, and $O$ denotes the relative dimensions of zero matrix.

Proof Taking $k = 0, 1, ...$ in system (2.20) and equations (3.3), we generate the following sequence

$\begin{eqnarray} \begin{array}{lll} y_0 = C_0x_0, \; \; \; y_1 = C_1x_1, \\ y_2 = C_2x_2 = C_2(\hat{A}_0x_0+\hat{A}_1x_1+\hat{B}_1u_1) = C_2\hat{A}_0x_0+C_2\hat{A}_1x_1+C_2\hat{B}_1u_1, \\ y_3 = C_3x_3 = C_3\hat{A}_1\hat{A}_0x_0+C_3(\hat{A}_0+\hat{A}^2_1)x_1+C_3\hat{A}_1\hat{B}_1u_1+C_3\hat{B}_2u_2. \end{array} \end{eqnarray}$

Then the measurement $y_k$ according to the equation (3.4) of state vector $x_k$ is

$\begin{equation} y_k = C_kQ(k)x_0+C_kP(k)x_1+\sum\limits_{i = 1}^{k-1}C_kM_i^{(k)}\hat{B}_iu_i, \end{equation}$

(4.1)

In general, we have

$\begin{equation} y_{2n-1} = C_{2n-1}Q(2n-1)x_0+C_{2n-1}P(2n-1)x_1+\sum\limits_{i = 1}^{2n-2}C_{2n-1}M_i^{(2n-1)}\hat{B}_iu_i, \end{equation}$

(4.2)

As consequence, equation (4.2) can be rewritten in the following relevant matrix form

$\begin{eqnarray} \left[ \begin{array}{llllll} y_0\\ y_1\\ y_2-C_2\hat{B}_1u_1\\ y_3-C_3\hat{A}_1\hat{B}_1u_1-C_3\hat{B}_2u_2\\ ......\\ y_{2n-1}-\sum\limits_{i = 1}^{2n-2}C_{2n-1}M_i^{(2n-1)}\hat{B}_iu_i \end{array} \right] = \left[ \begin{array}{llllll} C_0&O\\ O&C_1\\ C_2\hat{A}_0&C_2\hat{A}_1\\ C_3\hat{A}_1\hat{A}_0&C_3(\hat{A}_0+\hat{A}^2_1)\\ ...&...\\ C_{2n-1}Q(2n-1)&C_{2n-1}P(2n-1) \end{array} \right]\left[ \begin{array}{ll} x_0\\ x_1 \end{array} \right]. \end{eqnarray}$

We know from linear algebra that the system of linear algebra equations with 2n unknowns, equation (4.1) has a unique solution $(x_0, x_1)$ if and only if the system matrix has rank $2n$ :

$\begin{eqnarray} {\rm{rank}}\; \left[ \begin{array}{llllll} C_0&O\\ O&C_1\\ C_2\hat{A}_0&C_2\hat{A}_1\\ C_3\hat{A}_1\hat{A}_0&C_3(\hat{A}_0+\hat{A}^2_1)\\ ...&...\\ C_{2n-1}Q(2n-1)&C_{2n-1}P(2n-1) \end{array} \right] = {\rm{2n}}. \end{eqnarray}$

(4.3)

The matrix in (4.3) we denote by $\mathbb{S}.$ Then the initial values $x_0, x_1$ are determined uniquely, if and only if ${\rm{rank}}\; \mathbb{S} = {\rm{2n}}.$

Analogously, a numerical example is given as follows to verify the validity of observability analysis.

Example.3 Considering the following system with discrete method ^[20]

$\begin{eqnarray} \begin{array}{ll} &\ddot{x}(t) = \left[ \begin{array}{cc} 2&1\\ 3&4 \end{array} \right]x(t)+\left[ \begin{array}{ll} 1\\ 2 \end{array} \right]u(t),\\ &y(t) = [1\; \; \; \; 3]x(t). \end{array} \end{eqnarray}$

(4.4)

The discrete system has the form

$\begin{eqnarray} \begin{array}{ll} x_{k+1} = \hat{A}_0x_{k-1}+\hat{A}_1x_k+\hat{B}u_k, \\ y_k = Cx_k \end{array} \end{eqnarray}$

(4.5)

where

$\begin{eqnarray} \hat{A}_0 = -I_2, \; \; \hat{A}_1 = 2I_2+\tau^2 A, \; \; \hat{B} = \tau^2 B, \; \; \; A = \left[ \begin{array}{cc} 2&1\\ 3&4 \end{array} \right], \; \; B = \left[ \begin{array}{ll} 1\\ 2 \end{array} \right], \; \; C = [1\; \; \; \; 3]. \end{eqnarray}$

(4.6)

Letting $\tau = 0.1s,$ according to Theorem $2,$ we have

$\begin{eqnarray} n = 2, \; \hat{A}_1 = \left[ \begin{array}{cc} 2.02&0.01\\ 0.03&2.04 \end{array} \right], \; \hat{A}_1^2 = \left[ \begin{array}{cc} 4.0807&0.0406\\ 0.1218&4.1619 \end{array} \right], \end{eqnarray}$

(4.7)

$\begin{eqnarray} \mathbb{S} = \left[ \begin{array}{llll} C&O\\ O&C\\ C\hat{A}_0&C\hat{A}_1\\ C\hat{A}_1\hat{A}_0&C(\hat{A}_0+\hat{A}^2_1) \end{array} \right] = \left[ \begin{array}{llll} 1&3&0&0\\ 0&0&1&3\\ -1&-3&2.11&6.13\\ -2.11&-6.13&4.4461&12.5263 \end{array} \right], \end{eqnarray}$

(4.8)

$\begin{eqnarray} det\mathbb{S} = -0.04\neq 0, \; \; rank\mathbb{S} = 4 = 2n. \end{eqnarray}$

(4.9)

Therefore, the discrete system (4.5) is observable.

4.2. Numerical calculations of observability of discrete magnetic attitude control system

To system (2.19), according to equations (3.5), we also have

$\begin{eqnarray} \begin{array}{llllllll} &Q(2) = \hat{A}_0Q(0)+\hat{A}_1Q(1) = \hat{A}_0, \\ &Q(3) = \hat{A}_0Q(1)+\hat{A}_1Q(2) = \hat{A}_1\hat{A}_0, \\ &Q(4) = \hat{A}_0Q(2)+\hat{A}_1Q(3) = \hat{A}^2_0+\hat{A}^2_1\hat{A}_0, \\ &Q(5) = \hat{A}_0Q(3)+\hat{A}_1Q(4) = \hat{A}_0\hat{A}_1\hat{A}_0+\hat{A}_1(\hat{A}^2_0+\hat{A}^2_1\hat{A}_0), \\ &P(2) = \hat{A}_0P(0)+\hat{A}_1P(1) = \hat{A}_1, \\ &P(3) = \hat{A}_0P(1)+\hat{A}_1P(2) = \hat{A}_0+\hat{A}^2_1, \\ &P(4) = \hat{A}_0P(2)+\hat{A}_1P(3) = \hat{A}_0\hat{A}_1+\hat{A}_1\hat{A}_0+\hat{A}^3_1, \\ &P(5) = \hat{A}_0P(3)+\hat{A}_1P(4) = \hat{A}_0(\hat{A}_0+\hat{A}_1\hat{A}_0)+\hat{A}_1(\hat{A}_0\hat{A}_1+\hat{A}_1\hat{A}_0+\hat{A}^3_1), \end{array} \end{eqnarray}$

(4.10)

and

$\begin{eqnarray} \hat{A}_0 = \widetilde{A}_0 = \left[ \begin{array}{ccc} -1&0&-d_1\omega_0\tau \\ 0&-1&0\\ d_3\omega_0\tau&0&-1 \end{array} \right], \end{eqnarray}$

$\begin{eqnarray} \hat{A}_1 = \widetilde{A}_1 = \left[ \begin{array}{ccc} k_1\omega^2_0\tau^2+2&0&d_1\omega_0\tau \\ 0&k_2\omega_0^2 \tau^2+2&0\\ -d_3\omega_0\tau&0&k_3\omega^2_0\tau^2+2 \end{array} \right], \end{eqnarray}$

and we can rewrite the measurement matrix $C_k$ as the following form

$\begin{eqnarray} C_k = \left[ \begin{array}{lll} \alpha_2&0&-\alpha_1\sin\omega_0k\tau-\alpha_3\cos\omega_0k\tau\\ 0&-\alpha_2&\alpha_1\cos\omega_0k\tau-\alpha_3\sin\omega_0k\tau \end{array} \right], \; \; k = 0, 1, ..., 5 \end{eqnarray}$

letting

$\begin{eqnarray} -\alpha_1\sin\omega_0k\tau-\alpha_3\cos\omega_0k\tau = \sqrt{\alpha_1^2+\alpha_3^2}\sin(\omega_0k\tau+\phi)\triangleq{\eta_k(\phi)}, \\ \alpha_1\cos\omega_0k\tau-\alpha_3\sin\omega_0k\tau = \sqrt{\alpha_1^2+\alpha_3^2}\sin(\omega_0k\tau+\varphi)\triangleq{\eta_k(\varphi)}, \end{eqnarray}$

where

$\begin{eqnarray} \phi = \arctan{\frac{\alpha_3}{\alpha_1}}, \\ \varphi = \arctan{(-\frac{\alpha_1}{\alpha_3})}, \end{eqnarray}$

we get

$\begin{eqnarray} C_0 = \left[ \begin{array}{lll} \alpha_2&0&-\alpha_3\\ 0&-\alpha_2&\alpha_1 \end{array} \right], \; \; C_1 = \left[ \begin{array}{lll} \alpha_2&0&-\eta_1(\phi)\\ 0&-\alpha_2&\eta_1(\varphi) \end{array} \right], \; \; C_2 = \left[ \begin{array}{llllll} \alpha_2&0&-\eta_2(\phi)\\ 0&-\alpha_2&\eta_2(\varphi) \end{array} \right], \end{eqnarray}$

and

$\begin{eqnarray} C_2Q(2) = \left[ \begin{array}{lll} u_{51}&0&u_{53}\\ u_{61}&\alpha_2&u_{63} \end{array} \right], \; \; \; C_2P(2) = \left[ \begin{array}{lll} u_{54}&0&u_{56}\\ u_{64}&u_{65}&u_{66} \end{array} \right], \end{eqnarray}$

where

$\begin{eqnarray} \begin{array}{lllllllll} &u_{51} = -\alpha_2+d_3\omega_0\tau\eta_2(\phi), \\ &u_{53} = -\alpha_2d_1\omega_0\tau-\eta_2(\phi), \\ &u_{61} = d_3\omega_0\tau\eta_2(\varphi), \\ &u_{63} = -\eta_2(\varphi), \\ &u_{54} = \alpha_2(k_1\omega_0^2\tau^2+2)-d_3\omega_0\tau\eta_2(\phi), \\ &u_{56} = \alpha_2d_1\omega_0\tau+(k_3\omega_0^2\tau^2+2)\eta_2(\phi), \\ &u_{64} = -d_3\omega_0\tau\eta_2(\varphi), \\ &u_{65} = -\alpha_2(k_2\omega_0^2\tau^2+2), \\ &u_{66} = (k_3\omega_0^2\tau^2+2)\eta_2(\varphi). \end{array} \end{eqnarray}$

The following matrix $U$ is chosen as the first six columns of the observability matrix $\mathbb{S}$

$\begin{eqnarray} U = \left[ \begin{array}{llllll} \alpha_2&0&-\alpha_3&0&0&0\\ 0&-\alpha_2&\alpha_1&0&0&0\\ 0&0&0&\alpha_2&0&\eta_1(\phi)\\ 0&0&0&0&-\alpha_2&\eta_1(\varphi)\\ u_{51}&0&u_{53}&u_{54}&0&u_{56}\\ u_{61}&\alpha_2&u_{63}&u_{64}&u_{65}&u_{66} \end{array} \right]. \end{eqnarray}$

(4.11)

The determination of matrix $U$ is

$\begin{eqnarray} (-\alpha^2_2)(\alpha_2(u_{53}(u_{66}+(k_2\omega^2_0\tau^2+2)\eta_1(\varphi))-u_{56}(\alpha_1+u_{63}))+\eta_1(\phi)(u_{53}u_{64}-u_{54}(\alpha_1+u_{63}))\\ +\alpha_3((-\alpha_2)(u_{51}(u_{66}+(k_2\omega^2_0\tau^2)\eta_1(\varphi)-u_{56}u_{61})+\eta_1(\phi)(u_{51}u_{64}-u_{61}u_{54})))). \end{eqnarray}$

Generally speaking, the determination of matrix $U$ is not equal to 0 if and only if $\alpha_j\; (j = 1, 2, 3)\neq 0,$ which is valid according to the definition of $\alpha_j.$ In addition, using Matlab, choosing $\tau = 0.1,$ we also have

$\begin{eqnarray} {\rm{rank}}\; {\rm{U}} = \; {\rm{rank}}\; \mathbb{S} = 6. \end{eqnarray}$

(4.12)

Based on Theorem $2$ , the system (2.19) with backward Euler method is observable.

Moreover, we make $\tau = 0.2, \; 0.05$ respectivly to compute the rank of $\mathbb{S}$ and the results show different $\tau$ have no effect on the observability of system (2.19). The observability of system (2.14) with forward Euler method can be proven similarly.

5. Conclusions

In this paper, two different discrete schemes of the second-order linear time-varying system represented by the linearized satellite magnetic attitude control motion equation are obtained by Euler method. Subsequently, the controllability and observability conditions of a new discrete second-order linear time-varying system are proposed, which are applied to investigate the controllability and observability of the discretized satellite magnetic control system. Some numerical examples are given to further verify the correctness of theoretical results. Research results show that, generally speaking, different periods $\tau$ and parameters in coefficient matrices have no effect on the controllability and observability of the resulting discrete system.

Acknowledgments

This work was supported in part by the National Natural Science Foundation of China (12171196).

Conflict of interest

All authors declare no conflicts of interest in this paper.

References

[1]	M. Reyhanoglu, J. R. Hervas, Three-axis magnetic attitude control algorithms for small satellites, Proceedings of 5th International Conference on Recent Advances in Space Technologies, (2011), 897–902. https://dx.doi.org/10.1109/RAST.2011.5966973 doi: 10.1109/RAST.2011.5966973
[2]	A. L. Rodriquez-Vazquez, M. A. Martin-Prats, F. Bernelli-Zazzera, Spacecraft magnetic attitude control using approximating sequence Riccati equations, IEEE T. Aero. Elec. Sys., 51 (2015), 3374–3385. https://dx.doi.org/10.1109/TAES.2015.130478 doi: 10.1109/TAES.2015.130478
[3]	G. Dipak, M. Bijoy, T. N. Bidul, S. Manoranjan, Three-axis global magnetic attitude control of Earth-pointing satellites in circular orbit, Asian J. control, 19 (2017), 2028–2041. https://doi.org/10.1002/asjc.1506 doi: 10.1002/asjc.1506
[4]	E. I. Ergin, P. C. Wheeler, Magnetic control of a spinning satellite, J. Spacecraft Rockets, 2 (1965), 846–850. https://doi.org/10.2514/3.28302 doi: 10.2514/3.28302
[5]	M. L. Renard, Command laws for magnetic attitude control of spin-stabilized earth satellites, J. Spacecraft Rockets, 4 (1967), 156–163. https://doi.org/10.2514/3.28828 doi: 10.2514/3.28828
[6]	J. Kim, Y. Jung, H. Bang, Linear time-varying model predictive control of magnetically actuated satellites in elliptic orbits, Acta Astronaut., 151 (2018), 791–804. https://doi.org/10.1016/j.actaastro.2018.07.029 doi: 10.1016/j.actaastro.2018.07.029
[7]	R. Q. Chai, A. TsourdosH, J. Gao, Y. Q. Xia, S. C. Chai, Dual-Loop Tube-Based Robust Model Predictive Attitude Tracking Control for Spacecraft With System Constraints and Additive Disturbances, IEEE T. Ind. Electron., 69 (2022), 4022–4033. https://doi.org/10.1109/TIE.2021.3076729 doi: 10.1109/TIE.2021.3076729
[8]	H. Y. Yang, Q. L. Hu, H. Y. Dong, X. W. Zhao, ADP-Based Spacecraft Attitude Control Under Actuator Misalignment and Pointing Constraints, IEEE T. Ind. Electron., 69 (2022), 9342–9352. https://doi.org/10.1109/TIE.2021.3116571 doi: 10.1109/TIE.2021.3116571
[9]	K. Zhou, H. Huang, X. Wang, L. Sun, Magnetic attitude control for Earth-pointing satellites in the presence of gravity gradient , Aerosp. Sci. Technol., 60 (2017), 115–123. https://doi.org/10.1016/j.ast.2016.11.003 doi: 10.1016/j.ast.2016.11.003
[10]	V. V. Beletskii, Dvizhenie iskusstvennogo sputnika otnositel'no tsentra mass(Artificial satellite motion relative to center of mass), Moscow: Nauka, 1965.
[11]	M. L. Psiaki, Magnetic torquer attitude control via asymptotic periodic linear quadratic regulation, J. Guid. Control Dynam., 24 (2001), 386–394. https://doi.org/10.2514/2.4723 doi: 10.2514/2.4723
[12]	S. P. Bhat, Controllability of nonlinear time-varying systems: Applications to spacecraft attitude control using magnetic acutation, IEEE T. Automat. Contr., 50 (2005), 1725–1735. https://doi.org/0.1109/TAC.2005.858686
[13]	R. Sutherland, I. Kolmanovsky, A. R. Girard, Attitude control of a 2U cubesat by magnetic and air drag torques, IEEE T. Contr. Syst. T., 27 (2018), 1047–1059. https://doi.org/10.1109/TCST.2018.2791979 doi: 10.1109/TCST.2018.2791979
[14]	Y. Yang, Controllability of spacecraft using only magnetic torques, IEEE T. Aero. Elec. Sys., 52 (2016), 954–961. https://doi.org/10.1109/TAES.2015.150520 doi: 10.1109/TAES.2015.150520
[15]	V. M. Morozov, V. I. Kalenova, Satellite control using magnetic moments: Controllability and stabilization algorithms, Cosmic Research, 58 (2020), 158–166. https://doi.org/10.1134/S0010952520030041 doi: 10.1134/S0010952520030041
[16]	Y. Q. Wang, D. Zhao, Y. Y. Li, S. X. Ding, Unbiased Minimum Variance Fault and State Estimation for Linear Discrete Time-Varying Two-Dimensional Systems, IEEE T. Automat. Contr., 62 (2017), 5463–5469. https://doi.org/10.1109/TAC.2017.2697210 doi: 10.1109/TAC.2017.2697210
[17]	D. Zhao, H. K. Lam, Y. Y. Li, S. X. Ding, S. Liu, A Novel Approach to State and Unknown Input Estimation for Takagi-Sugeno Fuzzy Models With Applications to Fault Detection, IEEE Transactions on Circuits and Systems-I: Regular Papers, 67 (2020), 2053–2063. https://doi.org/10.1109/TCSI.2020.2968732 doi: 10.1109/TCSI.2020.2968732
[18]	Y. Y. Li, M. Yuan, M. Chadli, Z. P. Wang, D. Zhao, Unknown Input Functional Observer Design for Discrete-Time Interval Type-2 Takagi-Sugeno Fuzzy Systems, IEEE T. Fuzzy Syst., 30 (2022), 4690–4701. https://doi.org/10.1109/TFUZZ.2022.3156735 doi: 10.1109/TFUZZ.2022.3156735
[19]	G. Reissig, C. Hartung, F. Svaricek, Strong stuctural controllability and observability of linear time-varying systems, IEEE T. Automat. Contr., 59 (2014), 3087–3092. https://doi.org/10.1109/TAC.2014.2320297 doi: 10.1109/TAC.2014.2320297
[20]	E. N. Mahmudov, Controllability and observability of second order linear time invariant systems, Optimization and Control, 39 (2019), 16–25.
[21]	A. Babiarz, A. Czornik, S. Siegmund, On Stabilization of discrete time-varying systems, SIAM J. Control Optim., 59 (2021), 242–266. https://doi.org/10.1137/20M1324326 doi: 10.1137/20M1324326
[22]	M. Witczak, V. Puig, D. Rotondp, P. Witczak, A necessary and sufficient condition for total observability of discrete-time linear time-varying systems, International Federation Of Automatic Control, 50 (2017), 729–734. https://doi.org/10.1016/j.ifacol.2017.08.232 doi: 10.1016/j.ifacol.2017.08.232
[23]	E. N. Mahmudov, Second order discrete time-varying and time-invariant linear continuous systems and kalman type conditions, Numerical Algebra, Control And Optimization, 12 (2022), 353–371. https://doi.org/10.3934/naco.2021010 doi: 10.3934/naco.2021010
[24]	G. D. Meena, S. Janardhanan, Discretization of Linear Time-Varying Systems, 2020 International Conference on Emerging Frontiers in Electrical and Electronic Technologies (ICEFEET), (2020), 1–6.
[25]	V. I. Kalenova, V. M. Morozov, The reducibility of linear second-order time-varying systems with control and observation, Journal of Applied Mathematics and Mechanic, 76 (2012), 413–422. https://doi.org/10.1016/j.jappmathmech.2012.09.008 doi: 10.1016/j.jappmathmech.2012.09.008
[26]	C. Liu, C. Li, Reachability and observability of switched linear systems with continuous-time and discrete-time subsystems, International Journal Of Control, Automation And Systems, 11 (2013), 200–205. https://doi.org/10.1007/s12555-011-0169-4 doi: 10.1007/s12555-011-0169-4
[27]	D. S. Ivanov, M. Yu. Ovchinnikov, V. I. Penkov, D. S. Roldugin, D. M. Doronin, A. V. Ovchinnikov, Advanced numerical study of the three-axis magnetic attitude control and determination with uncertainties, Acta Astronaut., 132 (2017), 103–110. https://doi.org/10.1016/j.actaastro.2016.11.045 doi: 10.1016/j.actaastro.2016.11.045
[28]	M. G. Frost, Controllability, observability and the transfer function matrix for a delay-differential system, Int. J. Control, 35 (1982), 175–182. https://doi.org/10.1080/00207178208922610 doi: 10.1080/00207178208922610

This article has been cited by:

1.	Uliana Vladimirovna Monakhova, Orbital stabilization of dynamically elongated small satellite using active magnetic attitude control, 2024, 20712898, 1, 10.20948/prepr-2024-5
2.	Dmitry Roldugin, Anna Okhitina, Uliana Monakhova, Mikhail Ovchinnikov, Comparison of Feedback Three-Axis Magnetic Attitude Control Strategies, 2023, 10, 2226-4310, 975, 10.3390/aerospace10120975

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

AIMS Mathematics

1.8 3.4

Metrics

Article views(1244) PDF downloads(85) Cited by(2)

Preview PDF

Download XML

Export Citation

Article outline

Show full outline

Figures and Tables

Tables(1)

AIMS Mathematics

Controllability and observability of discretized satellite magnetic attitude control system

Related Papers:

Abstract

1. Introduction

2. Transformation of satellite attitude control system

2.1. Linearized equation of satellite magnetic control

2.2. Discretization of the linearized satellite control system

3. Controllability of the discrete system

3.1. Controllability analysis

3.2. Numerical calculations of controllability of discrete magnetic attitude control system

4. Observability of the discrete system

4.1. Observability analysis

4.2. Numerical calculations of observability of discrete magnetic attitude control system

5. Conclusions

Acknowledgments

Conflict of interest

References

This article has been cited by:

Reader Comments

通讯作者: 陈斌, bchen63@163.com

Metrics

Figures and Tables

Other Articles By Authors

Catalog

AIMS Mathematics

Controllability and observability of discretized satellite magnetic attitude control system

Related Papers:

Abstract

1. Introduction

2. Transformation of satellite attitude control system

2.1. Linearized equation of satellite magnetic control

2.2. Discretization of the linearized satellite control system

3. Controllability of the discrete system

3.1. Controllability analysis

3.2. Numerical calculations of controllability of discrete magnetic attitude control system

4. Observability of the discrete system

4.1. Observability analysis

4.2. Numerical calculations of observability of discrete magnetic attitude control system

5. Conclusions

Acknowledgments

Conflict of interest

References

This article has been cited by:

Reader Comments

通讯作者: 陈斌, bchen63@163.com

Metrics

Figures and Tables

Other Articles By Authors

Related pages

Tools

Export File

Citation

Format

Content

Catalog