- Research
- Open access
- Published:
Exponentially stable solution of mathematical model of agents dynamics on time scales
Advances in Difference Equations volume 2019, Article number: 231 (2019)
Abstract
This work is an attempt at studying leader-following model on the arbitrary time scale. The step size is treated as a function of time. Our purpose is establishing conditions ensuring a leader-following consensus for any time scale basing on the Grönwall inequality. We give some examples illustrating the obtained results.
1 Introduction
The aim of this paper is to propose conditions ensuring the consensus of a multi-agent system over an arbitrary time scale. We consider continuous- and discrete-time models and also models on time scales simultaneously consisting of both kinds of points, right-dense and right-scattered. Under some assumptions, we prove that consensus can be achieved exponentially if the graininess function is bounded. All theorems are still true if the graininess function approaches zero. Some existing results of discrete-time consensus are particular cases of the results presented in this paper.
The leader-following problem has been investigated since 1970s. In 1974, DeGroot [1] studied an explicitly described model that resulted in the consensus. In 2000, Krause [2, 3] discussed the model of a group of agents who have to make a joint assessment of a certain magnitude. The coordination of groups of mobile autonomous agents based on the nearest-neighbor rules was considered by Jadbabaie et al. [4]. Blondel et al. [5, 6], took into account Krause’s model with state-dependent connectivity. Girejko et al. [7, 8], examined Krause’s model on discrete time scales. In 2007 there were published two important papers by Cucker and Smale [9, 10]. The authors considered an emergent behavior in flocks. The Cucker–Smale model on isolated time scales was in the area of interests of Girejko et al. [8]. Girejko, Machado, Malinowska, and Martins [11] investigated sufficient conditions for consensus in the Cucker–Smale-type model on discrete time scales. In 2015, Wang et al. [12] published some results for the leader-following consensus of discrete-time linear multi-agent systems with communication noises.
Our results generalize and improve the results obtained in [13] and [14]. In [14] consensus on different types of discrete-time scales is considered under the assumption that the feedback control gain γ is constant.
2 Basis of time scales calculus
A time scale is a model of time [15,16,17], where the step size is a function of time. From mathematical point of view, it is an arbitrary nonempty closed subset \(\mathbb{T}\) of the set \(\mathbb{R}\) of real numbers.
The mapping \(\sigma :\mathbb{T}\to \mathbb{T}\) defined by \(\sigma (t)= \inf {\{s\in \mathbb{T}:s>t\}}\) with \(\inf \emptyset =\sup \mathbb{T}\) is called the forward jump operator. Similarly, we define the backward jump operator \(\rho :\mathbb{T}\to \mathbb{T}\) by \(\rho (t)=\sup {\{s\in \mathbb{T}:s< t\}}\) with \(\sup \emptyset =\inf \mathbb{T}\). The following classification of points is used within the theory: a point \(t\in \mathbb{T}\) is called right-dense, right-scattered, left-dense, and left-scattered if \(\sigma (t)=t\) (for \(t < \sup \mathbb{T}\)), \(\sigma (t)>t\), \(\rho (t)=t\) (for \(t > \inf \mathbb{T}\)), and \(\rho (t)< t\), respectively. We say that t is isolated if \(\rho (t)< t<\sigma (t)\) and that t is dense if \(\rho (t)=t=\sigma (t)\). The function \(\mu :\mathbb{T}\to [0, \infty )\) is defined by \(\mu (t)=\sigma (t)-t\) and called the graininess function. The delta (or Hilger) derivative of \(f :\mathbb{T} \to \mathbb{R}\) at a point \(t \in \mathbb{T}^{\kappa }\), where
is defined in the following way.
Definition 1
([16])
The delta derivative \(f^{\Delta }(t)\) is the number (provided that it exists) with the property that given any \(\varepsilon >0\), there is a neighborhood U of t (i.e., \(U=(t- \delta , t+ \delta ) \cap \mathbb{T}\) for some \(\delta >0\)) such that
Moreover, we say that f is delta (or Hilger) differentiable (or, shortly, differentiable) on \(\mathbb{T}^{\kappa }\) if \(f^{\Delta }(t)\) exists for all \(t\in \mathbb{T}^{\kappa }\). Then the function \(f^{\Delta }:\mathbb{T}^{\kappa }\to \mathbb{R}\) is called the (delta) derivative of f on \(\mathbb{T}^{\kappa }\).
The following definitions will be further used, too.
Definition 2
([16])
A function \(f :\mathbb{T} \to \mathbb{R}\) is called regulated if its right-sided limits exist (finite) at right-dense points in \(\mathbb{T}\) and its left-sided limits exist (finite) at left-dense points in \(\mathbb{T}\). A function \(f :\mathbb{T} \to \mathbb{R}\) is called rd-continuous if it is continuous at right-dense points in \(\mathbb{T}\) and its left-side limits exist (finite) at left-dense points in \(\mathbb{T}\).
Definition 3
([16])
Let \(f :\mathbb{T} \to \mathbb{R}\) be a regulated function. We define the indefinite integral of f by \(\int f(t) \Delta t = F(t) +c\), where c is an arbitrary constant, and F is a preantiderivative of f. We define the Cauchy integral of f as \(\int _{a}^{b} f(t) \Delta t = F(b) - F(a)\) for all \(a,b \in \mathbb{T}\).
Definition 4
([16])
We say that a function \(p :\mathbb{T} \to \mathbb{R}\) is regressive if \(1+ \mu (t)p(t) \neq 0\) for all \(t \in \mathbb{T}^{ \kappa }\). The set of all regressive and rd-continuous functions \(p :\mathbb{T} \to \mathbb{R}\) is denoted by \(\mathcal{R}\). The set of all positively regressive elements of \(\mathcal{R}\) is defined as \(\mathcal{R}^{+} := \{p \in \mathcal{R} :1+ \mu (t)p(t) > 0 \mbox{ for all } t \in \mathbb{T} \}\).
Definition 5
([16])
An \(N \times N\)-matrix-valued function P on a time scale \(\mathbb{T}\) is called regressive (with respect to \(\mathbb{T}\)) if
where by I we denote the \(N \times N\) identity matrix.
Similarly to the scalar case, the class of all regressive and rd-continuous matrix-valued functions is denoted by \(\mathcal{R}\). Notice that a constant \(N \times N\) matrix P is regressive iff the eigenvalues \(\lambda _{i}\) of P are regressive for all \(1 \leq i \leq N\).
We use the Grönwall inequality in the proof of the main result.
Lemma 1
([16])
Let z be rd-continuous, \(p \in \mathcal{R}^{+}\), and \(p(t)\geq 0\) for \(t \in \mathbb{T}\) and \(c \in \mathbb{R}\). Then
implies
Here \(z(t)=e_{p}(t,T_{0})\), \(T_{0} \in \mathbb{T}\), is a solution of the initial value problem
We assume that
which implies that \(\mathbb{T}^{\kappa }= \mathbb{T}\).
3 Mathematical model of agents dynamics
We consider a multi-agent system consisting of N agents and the leader. The dynamics of each agent labeled i, \(i= 1,2, \ldots, N\), is given by the following equation:
where \(t \in \mathbb{T}\), \(x_{i} :\mathbb{T} \to \mathbb{R}\), and \(\gamma :\mathbb{T} \to \mathbb{R}\) represent the state and the feedback control gain at time t, respectively. Here \(a_{ij} \in \mathbb{R}\), \(d_{i} \in \mathbb{R}\), \(i,j=1,2, \ldots ,N\), and \(D := \operatorname{diag}[d_{1},d_{2}, \ldots , d_{N}]\) is a diagonal matrix. Throughout this paper, we assume that \(a_{ij}=a_{ji}\), which means that the matrix \(A=[a_{ij}]_{N \times N}\) is symmetric. The function \(f:\mathbb{T} \times \mathbb{R} \to \mathbb{R}\) describes nonlinear dynamics. The leader, labeled as \(i=0\), for multi-agent system (2) is an isolated agent with trajectory described by
Notice that the control law \(\gamma (t) \sum_{j=1}^{N}a_{ij}(x _{j}(t)-x_{i}(t))+\gamma (t) d_{i}(x_{0}(t)-x_{i}(t))\) for the ith agent used in system (2)–(3) was studied by many authors, including Yu, Jiang, and Hu [18].
Let \(\varepsilon _{i}(t)=x_{i}(t)-x_{0}(t)\) denote the distance between the leader and the ith agent. From (2)–(3) we obtain
for \(i=1,2, \ldots , N\). Setting
and
system (2)–(3) takes the following form:
(for details, see [19]). Here is the vector \((1,\ldots ,1)^{T}\). Recall that \(B = L+D\) is a symmetric matrix since A is a symmetric matrix and L is the Laplacian matrix \(L=[l_{ij}]\) with \(l_{ii}=\sum_{j \neq i}a_{ij}\) and \(l_{ij}=-a_{ij}\), \(i,j=1,\ldots ,N\), \(i\neq j\).
If
in equation (1), then by \(e_{-\gamma B}(t,T_{0})\) we denote a solution of the initial value problem
By variation of constants (see [16]) the unique solution of equation (5) is given by
Definition 6
A function \(F :\mathbb{T}\times \mathbb{R}^{N} \to \mathbb{R} ^{N}\) fulfills Lipschitz condition with respect to the second variable if there exists a positive constant \(\mathcal{L}\) such that
Definition 7
We say that equation (5), where \(T_{0} \geq 0\), \(\varepsilon _{T_{0}} \in \mathbb{R}^{N}\), is exponentially stable if there exist positive constants c and d such that, for any rd-continuous solution \(\varepsilon (t, T_{0}, \varepsilon _{T_{0}})\) of equation (5), we have
For some relevant result on the exponential stability in the discrete case, see [20] and [21].
Definition 8
The multi-agent system (2)–(3) is said to achieve the leader-following consensus exponentially if equation (5) is exponentially stable.
In 2005, Peterson and Raffoul [22] investigated the exponential stability of the zero solution to systems of dynamic equations on time scales. The authors defined suitable Lyapunov-type functions and then formulated appropriate inequalities on these functions that guarantee that the zero solution exponentially decays to zero. For the growth of generalized exponential functions on time scales, see Bodine and Lutz [23].
4 Main results
Assume that the function \(F :\mathbb{T}\times \mathbb{R}^{N} \to \mathbb{R}^{N}\) defined by (4) satisfies Lipschitz condition with respect to the second variable.
Let \(\lambda _{i}\), \(i=1,2,\ldots ,N\), be the eigenvalues of the matrix B. By \(\mathbb{T}^{s}\) and \(\mathbb{T}^{d}\) we denote the sets of right-scattered and right-dense points of \(\mathbb{T}\), respectively. Notice that since by assumption \(\sup \mathbb{T}= \infty \), at least one of the sets \(\mathbb{T}^{s}\) or \(\mathbb{T}^{d}\) must be unbounded.
Next, we rewrite the time scale \(\mathbb{T}\) in a useful way for estimation of the norm of a solution of the initial value problem (5) on a time scale consisting of right-scattered and right-dense points. To avoid confusion, we underline that any interval throughout this paper is an interval on the time scale, that is, any interval contains only points of the time scale. Set
for \(i=1,2, \ldots \) . In case of \([T_{2i-1}, \infty ) \cap \mathbb{T} ^{s} = \emptyset \) for some \(i \in \mathbb{N}\) we take \(T_{2i}=\infty \) (see Example 9). If \([T_{2i-1}, t) \cap \mathbb{T}^{d} = [T _{2i-1}, T_{2i-1}) =\emptyset \) for some \(i \in \mathbb{N}\), then we also take \(T_{2i}=\infty \) (see Example 6). Analogously, if \([T_{2i}, \infty ) \cap \mathbb{T}^{d} = \emptyset \) for some \(i \in \mathbb{N}\), then \(T_{2i+1}=\infty \). If \(T_{j} = \infty \) for some \(j \in \mathbb{N}_{0}\), then we take \(T_{j+i}=T_{j}\) for \(i \in \mathbb{N}\) and \([T_{j+i}, T_{j+i+1}) = (T_{j+i}, T_{j+i+1}]= \emptyset \) (see Example 7). We see that if \(\sigma (T_{0})=T _{0}\), then \(T_{1}=T_{0}\).
Example 1
Let \(\mathbb{T}= \{1\} \cup [2,3] \cup [6, \infty )\). Here \(T_{0}=1\), \(T_{1}=2\), \(T_{2}=3\), \(T_{3}=6\), and \(T_{4}=\infty \).
We underline that \(T_{2i+1} \in \mathbb{T}^{d}\) for any \(i \in \mathbb{N}_{0}\), but it is possible that \(T_{2i} \notin \mathbb{T} ^{s}\) for some \(i \in \mathbb{N}_{0}\).
Example 2
Let \(\mathbb{T}= \bigcup_{i=1}^{\infty }[2i-1, 2i] \cup \{4i + \frac{1}{j+1} :i, j \in \mathbb{N}\}\). Here \(T_{0}=1\), \(T_{1}=T_{0}\), and \(T_{i}=i\) for \(i \in \{2,3, \ldots \}\). We see that \(T_{1}=T_{0} \in \mathbb{T}^{d}\), \(T_{2i+1} \in \mathbb{T}^{d}\), \(T_{4i} \in \mathbb{T}^{d}\), and \(T_{4i+2} \in \mathbb{T}^{s}\) for \(i \in \mathbb{N}\).
Example 3
Let \(\mathbb{T}= \bigcup_{i=1}^{\infty }[2i-1, 2i] \cup \{4i+1 - \frac{1}{j+1} :i, j \in \mathbb{N}\}\). Here \(T_{0}=1\), \(T_{i}=i\) for \(i \in \mathbb{N}\), and \(T_{2i-1} \in \mathbb{T}^{d}\) and \(T_{2i} \in \mathbb{T}^{s}\) for \(i \in \mathbb{N}\).
We can write
wherein \((T_{2i}, T_{2i+1}) \subset \mathbb{T}^{s}\) and \((T_{2i+1}, T _{2i+2}) \subset \mathbb{T}^{d}\).
In the next lemma, for any \(i \in \mathbb{N}\), we present the estimates of the norms of matrices \(e_{-\gamma B}(t,T_{2i}) \) for \(t \in [T_{2i}, T_{2i+1})\) and \(e_{-\gamma B}(t,T_{2i+1})\) for \(t \in [T_{2i+1}, T _{2i+2})\).
Lemma 2
Suppose that, for \(i = 1,2, \ldots , N\), the following conditions are satisfied:
Then there exists a positive real number \(\mathcal{M}<1\) such that
where \(\Vert \cdot \Vert \) denotes the spectral norm of the considered matrix at the point t.
Proof
Obviously, \(\mathbb{T}^{s} \cup \mathbb{T}^{d} =\mathbb{T}\). We consider two cases:
-
(i)
\(t\in \mathbb{T}^{s}\);
-
(ii)
\(t\in \mathbb{T}^{d}\).
In case (i), notice that since matrix B is symmetric, \(I- \mu (t) \gamma (t) B\) is also a symmetric matrix at the point t. Therefore \(\Vert I- \mu (t)\gamma (t) B \Vert \) equals the maximum of the absolute value of eigenvalues of the matrix \(I- \mu (t)\gamma (t) B\). This means that
for \(t \in [T_{2i}, T_{2i+1})\). Because of the positivity of μ on \(\mathbb{T}^{s}\) and condition (8), we have \(\vert \mu (s)\gamma (s) \lambda _{i} \vert = \mu (s)\gamma (s) \lambda _{i} \). Moreover, by (9), \(\mu (s)\gamma (s) \lambda _{i} \in (0,1)\) for \(i \in \{1,2, \ldots , N\}\). We can conclude
Again by (9), we have
By the preceding we have
where \(\mathcal{M}^{*} := 1-\delta \in (0,1)\).
Case (ii). Condition (8) implies
- (1∘):
-
\(\lambda _{i}>0\) for any \(i=1,2,\ldots , N\) and \(\gamma (t)>0\) for any \(t \in \mathbb{T}^{d}\)
or
- (2∘):
-
\(\lambda _{i}<0\) for any \(i=1,2,\ldots , N\) and \(\gamma (t)<0\) for any \(t \in \mathbb{T}^{d}\).
If (1∘), then
for \(t \in [T_{2i+1}, T_{2i+2})\), where \(\mathcal{M}^{**}:= ( \max_{i \in \{1,2, \ldots , N\}} \{e^{ \lambda _{i} }\})^{ - 1} \in (0,1)\).
If (2∘), then
for \(t \in [T_{2i+1}, T_{2i+2})\), where \(\mathcal{M}^{**}:= \max_{i \in \{1,2, \ldots , N\}} \{e^{ \lambda _{i} }\} \in (0,1)\).
Set \(\mathcal{M}:= \max \{\mathcal{M}^{*}, \mathcal{M}^{**}\}\). Obviously \(\mathcal{M}\in (0,1)\).
Hence \(\Vert e_{-\gamma B}(t,T_{2i}) \Vert \leq \prod_{s \in [T_{2i}, t)} \mathcal{M}\) for \(t \in [T_{2i}, T_{2i+1})\) and \(\Vert e_{-\gamma B}(t,T_{2i+1}) \Vert \leq \mathcal{M}^{ \int _{T_{2i+1}}^{t}\vert \gamma (s) \vert \,ds }\) for \(t \in [T_{2i+1}, T_{2i+2})\). □
Next, we find the estimates of the norm of the matrix \(e_{-\gamma B}(t,T _{0}) \) in two cases, \(t \in [T_{2i}, T_{2i+1})\) and \(t \in [T_{2i+1}, T_{2i+2})\).
Lemma 3
If conditions (8)–(9) are satisfied, then
for \(t \in [T_{2i}, T_{2i+1})\), and
for \(t \in [T_{2i+1}, T_{2i+2})\), where \(i \in \mathbb{N}_{0}\).
Proof
Let us rewrite the function \(e_{-\gamma B}(t,T_{0})\) in the form
or
By submultiplicativity of the norm, for \(t \in [T_{2i},T_{2i+1})\), we estimate the norm of the matrix \(e_{-\gamma B}(t,T_{0})\):
Analogously, for \(t \in [T_{2i+1},T_{2i+2})\), we get
□
Remark 1
If conditions (8)–(9) are satisfied, then
Proof
Since \(\mathcal{M} \in (0,1)\) and \(\sum_{j=1}^{i}{\int _{T_{2j-1}}^{T _{2j}}\vert \gamma (s) \vert \,ds}+\int _{T_{2i+1}}^{t}\vert \gamma (s) \vert \,ds \geq 0\), we get
From the above, inequalities (10) and (11) imply
where \(i \in \mathbb{N}_{0}\), and this is our claim. □
The following result concerns the scalar case of exponential function on arbitrary time scale.
Lemma 4
Let \(e_{\mathcal{L} \mathcal{M}^{-1}}( \cdot , T_{0}) :\mathbb{T} \to \mathbb{R}\). Then
for \(t \in [T_{2i}, T_{2i+1})\), and
for \(t \in [T_{2i+1}, T_{2i+2})\), where \(i \in \mathbb{N}_{0}\).
We are now in a position to present the main theorem of this paper.
Theorem 1
Suppose that conditions (7)–(9) are satisfied and for any \(t \in \mathbb{T}\),
and
Then equation (5) is exponentially stable.
Proof
Taking the norm of the both sides of equation (6), we obtain
Using properties of the norm, we get
and consequently
By condition (7) we obtain
For \(t \in [T_{2i}, T_{2i+1})\), using (10), we estimate
Multiplying both sides of this inequality by
we obtain
Since \(\mathcal{M}^{-\int _{T_{0}}^{T_{0}}\vert \gamma (s) \vert \,ds}=1\),
Using \(\sigma (\tau )= \tau \), we get
By Lemma 1 this leads to the inequality
Using Lemma 4, we have
Hence
By (12) we have
Analogously, for \(t \in (T_{2i+1}, T_{2i+2}] \),
Set
for \(i \in \mathbb{N}_{0}\), and
By (13) and (14) inequalities (15) and (16) imply the thesis. □
Corollary 1
Let conditions (7)–(9) and (12) be satisfied. Moreover, suppose that
and
Then equation (5) is exponentially stable.
Proof
By (17) we get that \(t \to \infty \) iff \(i \to \infty \). Since \(0<\mathcal{M}+ \mu ^{*}\mathcal{L} <1\) and \(\mathcal{M} \in (0,1)\), by properties of the functions \(\mathcal{M}^{t}\) and \(e^{t}\) condition (19) implies conditions (13) and (14). Hence assumptions of Theorem 1 are satisfied. So, the statement holds. □
Example 4
Let
Here \(\mathbb{T}^{d}=\bigcup_{i=3}^{\infty } [\frac{i}{2}, \frac{i}{2}+\frac{1}{i^{3}} )\) and \(\mathbb{T}^{s}=\{ \frac{i}{2}+\frac{1}{i^{3}} :i \in \mathbb{N}, i \geq 3\}\),
Moreover, let
Hence
We have \(\mathcal{L}= 0.100\), \(\lambda _{1}= 2 - \sqrt{2}\), \(\lambda _{2}=2\), \(\lambda _{3}= 3\), and \(\lambda _{4}= 2 + \sqrt{2}\). From this it follows that \(\lambda _{\min }=\min \{\lambda _{1}, \lambda _{2},\lambda _{3}, \lambda _{4}\}\approx 0.585\) and
From the above we get
and
All assumptions of Corollary 1 are satisfied, and thus equation (5) is exponentially stable. System (2)–(3) achieves consensus exponentially.
In Example 4 we have
but this condition is not required for the exponential stability of (5) (see Example 5).
Remark 2
If conditions (7)–(9), (12), and (18) are satisfied and if
and
then equation (5) is exponentially stable.
Proof
If condition (21) holds, then
By (22) we see that \(\operatorname{sum}(i)< 0\) for any \(i \in \mathbb{N}\) and \(e^{\operatorname{sum}(i)}\) is a positive decreasing function of the variable \(i\in \mathbb{N}\). Here \(\prod_{s \in \mathbb{T}^{s}}(\mathcal{M}+ \mu ^{*} \mathcal{L})\) and \(e^{\operatorname{sum}(i)}\) for any \(i \in \mathbb{N}\) are bounded. If the cardinality of set \(\mathbb{T}^{s}\) is infinity, then \(\lim_{s \to \infty }\prod_{s \in \mathbb{T}^{s}}( \mathcal{M}+ \mu ^{*} \mathcal{L})=0\). If the cardinality of set \(\mathbb{T}^{d}\) is infinity, then \(\lim_{i \to \infty }e^{ \operatorname{sum}(i)}=0\). Thus, by Theorem 1, we obtain the statement. □
Example 5
Let
Here \(\mathbb{T}^{d}=\bigcup_{i=3}^{\infty } [\frac{i}{2}, \frac{i}{2}+\frac{1}{i} )\) and \(\mathbb{T}^{s}=\{\frac{i}{2}+ \frac{1}{i} :i \in \mathbb{N}, i \geq 3\}\),
Moreover, let
and let the matrix B be given by (20) in equation (5). Then \(\mathcal{L}= 0.250\), \(\lambda _{\min } \approx 0.585\), and
Finally,
All assumptions of Remark 2 hold, and thus equation (5) is exponentially stable.
In Example 5 we have
although system (2)–(3) achieves consensus exponentially.
Corollary 2
If conditions (7)–(9) and (18) are satisfied and if
then equation (5) is exponentially stable.
Proof
Since (23) holds,
Hence, recalling that the cardinality of the set \(\mathbb{T}^{s}\) is infinity, by (18) we obtain
where \(c^{*}=e^{\operatorname{sum}(i)}\). □
For two possible cases of carrying out of assumption (23), see Examples 4 and 7.
Theorem 1 generalizes Theorem 2 [14]. In the following example we present an equation on time scale for which Theorem 2 [14] cannot be applied, but our Corollary 2 of Theorem 1 can be.
Example 6
Let
Here \(\mathbb{T}^{d}=\{i :i \in \mathbb{N}\}\) and \(\mathbb{T} ^{s}=\{i+\frac{1}{j+1} :i,j \in \mathbb{N}, j\geq 2\}\),
Set \(f(t,x)=0.250 x\),
and B is given by (20) in equation (5). We have \(\mathcal{L}= 0.250\), \(\lambda _{\min }= 0.585\), and
Hence
All assumptions of Corollary 2 are satisfied, and thus equation (5) is exponentially stable.
Since \(\liminf_{t \to \infty }\mu (t) = 0\), the results obtained in [14] cannot be applied.
The following examples show two different situations concerning time scale in which condition (23) is satisfied. In the first example, \(\mathbb{T}^{d}\) is a bounded set. In the second one, \(\mathbb{T}^{d}\) is unbounded.
Example 7
Let
Here \(\mathbb{T}^{d}=[1,2] \cup [3,7]\) is bounded, and \(\mathbb{T} ^{s}=\{n :n \in \mathbb{N}, n\geq 8 \}\). We see that
Let also
and let the matrix B be given by (20) in equation (5). Then \(\mathcal{L}= 0.250\), \(\lambda _{\min }= 0.585\) and
As a consequence,
All assumptions of Corollary 2 are satisfied, and thus equation (5) is exponentially stable. This means that the multi-agent system (2)–(3) achieves the leader-following consensus exponentially.
Example 8
Let
Here either \(\mathbb{T}^{d} = \bigcup_{i=3}^{\infty } [ \frac{i}{2}+\frac{1}{i+1}, \frac{i}{2}+\frac{1}{i})\) or \(\mathbb{T} ^{s}=\{ \frac{i}{2}+\frac{1}{i} :i \in \mathbb{N}, i \geq 3 \}\) is an unbounded set. We see that
Moreover,
and the matrix B is given by (20) in equation (5). Then \(\mathcal{L}= 0.250\), \(\lambda _{\min }= 0.585\), and
Hence
All assumptions of Corollary 2 are satisfied, and thus equation (5) is exponentially stable. System (2)–(3) achieves consensus exponentially.
Notice that in Example 8 we have
Remark 3
If conditions (7)–(9) are satisfied,
and
then equation (5) is exponentially stable.
(See Example 4.)
Remark 4
Let conditions (7)–(9) be satisfied. If the cardinality of the set \(\mathbb{T}^{s}\) is finite and \(sum(i)<0\) for any \(i \in \mathbb{N}\), then equation (5) is exponentially stable.
Example 9
Let
Here \(\mathbb{T}^{d}=[12, \infty )\) is an unbounded set, whereas \(\mathbb{T}^{s}=\{1\} \cup \{11\}\) is bounded, and
Let
and let the matrix B be given by (20) in equation (5). Then \(\mathcal{L}= 0.100\), \(\lambda _{\min } \approx 0.585\), and
Hence
All assumptions of Remark 4 are satisfied, and thus equation (5) is exponentially stable.
References
DeGroot, M.H.: Reaching a consensus. J. Am. Stat. Assoc. 69, 118–121 (1974). https://doi.org/10.1080/01621459
Krause, U.: A discrete nonlinear and non-autonomous model of consensus formation. In: Communications in Difference Equations. ICDEA, vol. 1998. Gordon & Breach, Poznań (2000). https://doi.org/10.1109/IYCE.2015.7180798
Hegselmann, R., Krause, U.: Opinion dynamics and bounded confidence: models, analysis, and simulation. J. Artif. Soc. Soc. Simul. 5, 1–33 (2002)
Jadbabaie, A., Lin, J., Morse, A.S.: Coordination of groups of mobile autonomous agents using nearest neighbor rules. IEEE Trans. Autom. Control 48(6), 988–1001 (2003). https://doi.org/10.1109/TAC.2003.812781
Blondel, V.D., Hendrickx, J.M., Tsitsikli, J.N.: On Krause’s multi-agent consensus model with state-dependent connectivity. IEEE Trans. Autom. Control 54(11), 2586–2597 (2009). https://doi.org/10.1109/TAC.2009.2031211
Blondel, V.D., Hendrickx, J.M., Tsitsikli, J.N.: Continuous-time average-preserving opinion dynamics with opinion-dependent communications. SIAM J. Control Optim. 18(8), 5214–5240 (2010). https://doi.org/10.1137/090766188
Girejko, E., Machado, L., Malinowska, A.B., Martins, N.: Krause’s model of opinion dynamics on isolated time scales. Math. Methods Appl. Sci. 39(18), 5302–5314 (2016). https://doi.org/10.1002/mma.3916
Girejko, E., Malinowska, A.B., Schmeidel, E., Zdanowicz, M.: The emergence on isolated time scales. In: 21st International Conference on Methods and Models in Automation and Robotics (MMAR), IEEExplore (2016). https://doi.org/10.1109/MMAR.2016.7575317
Cucker, F., Smale, S.: On the mathematics of emergence. Jpn. J. Math. 2(1), 197–227 (2007). https://doi.org/10.1007/s11537-007-0647-x
Cucker, F., Smale, S.: Emergent behavior in flocks. IEEE Trans. Autom. Control 52(7), 852–862 (2007). https://doi.org/10.1109/TAC.2007.895842
Girejko, E., Machado, L., Malinowska, A.B., Martins, N.: On consensus in the Cucker–Smale type model on isolated time scale. Discrete Contin. Dyn. Syst., Ser. B 11(1), 77–89 (2018). https://doi.org/10.3934/dcdss.2018005
Wang, Y., Cheng, L., Wang, H., Hou, Z.G., Tan, M., Yu, H.: Leader-following consensus of discrete-time linear multi-agent systems with communication noises. In: 34th Control Conference (CCC), Hangzhou, China. Lecture Notes in Electrical Engineering, vol. 407. IEEE, New York (2015). https://doi.org/10.1109/ChiCC.2015.7260748
Ostaszewska, U., Schmeidel, E., Zdanowicz, M.: Leader-following consensus on discrete time scales. In: ICNAAM 2017, Thessaloniki, Greece. AIP Conference Proceedings. American Institute of Physics, New York (2018). https://doi.org/10.1063/1.5044162
Ostaszewska, U., Schmeidel, E., Zdanowicz, M.: Emergence of consensus of multi-agents systems on time scales. Miskolc Math. Notes (to appear)
Aulbach, B., Hilger, S.: A unified approach to continuous and discrete dynamics. In: Qualitative Theory of Differential Equations (Szeged, 1988). Colloq. Math. Soc. Jámos Bolyai, vol. 53. North-Holland, Amsterdam (1990)
Bohner, M., Peterson, A.: Dynamic Equations on Time Scales. Birkhäuser, Boston (2001). https://doi.org/10.1109/IYCE.2015.7180798
Bohner, M., Peterson, A.: Advances in Dynamic Equations on Time Scales. Birkhäuser, Boston (2003)
Yu, Z., Jiang, H., Hu, C.: Leader-following consensus of fractional-order multi-agent systems under fixed topology. Neurocomputing 149, 613–620 (2015). https://doi.org/10.1016/j.neucom.2014.08.013
Schmeidel, E.: The existence of consensus of a leader-following problem with Caputo fractional derivative. Opusc. Math. 39(1), 77–89 (2019). https://doi.org/10.7494/OpMath.2019.39.1.77
Berezansky, L., Migda, M., Schmeidel, E.: Some stability conditions for scalar Volterra difference equations. Opusc. Math. 36(4), 459–470 (2016). https://doi.org/10.7494/OpMath.2016.36.4.459
Elaydi, S.N.: An Introduction to Difference Equations. Springer, New York (2005)
Peterson, A., Raffoul, Y.N.: Exponential stability of dynamic equations on time scales. Adv. Differ. Equ. 2005, 858671 (2005). https://doi.org/10.1155/ADE.2005.133
Bodine, S., Lutz, D.A.: Exponential functions on time scales: their asymptotic behavior and calculation. Dyn. Syst. Appl. 12(1–2), 23–43 (2003)
Acknowledgements
Not applicable.
Availability of data and materials
Not applicable.
Funding
Ewa Schmeidel was supported by the Polish National Science Center grant on the basis of decision DEC-2014/15/B/ST7/05270. Urszula Ostaszewska and Małgorzata Zdanowicz have been supported by the Polish Ministry of Science and Higher Education under a subsidy for maintaining the research potential of the Faculty of Mathematics and Informatics, University of Bialystok.
Author information
Authors and Affiliations
Contributions
The main idea of this paper comes from Ewa Schmeidel. Proofs of theorems and examples are the joint work of all coauthors. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare that they have no competing interests.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Schmeidel, E., Ostaszewska, U. & Zdanowicz, M. Exponentially stable solution of mathematical model of agents dynamics on time scales. Adv Differ Equ 2019, 231 (2019). https://doi.org/10.1186/s13662-019-2159-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s13662-019-2159-4