You are on page 1of 7
<a href=Systems & Control Letters 85 (2015) 77–83 Contents lists available at ScienceDirect Systems & Control Letters journal homepage: www.elsevier.com/locate/sysconle Stability analysis of large-scale distributed networked control systems with random communication delays: A switched system approach Kooktae Lee , Raktim Bhattacharya Department of Aerospace Engineering, Texas A&M University, College Station, TX 77843-3141, USA a r t i c l e i n f o Article history: Received 3 December 2014 Received in revised form 22 April 2015 Accepted 10 August 2015 Available online 18 October 2015 Keywords: Large-scale distributed networked control system Markov jump linear system Switched system Random communication delays a b s t r a c t In this paper, we consider the stability analysis of large-scale distributed networked control systems with random communication delays. The stability analysis is performed in the switched system framework, particularly as the Markov jump linear system. There have been considerable research on stability analysis of the Markov jump systems. However, these methods are not applicable to large-scale systems because large numbers of subsystems result in extremely large number of switching modes. To circumvent this scalability issue, we propose a new reduced mode model for stability analysis, which is computationally scalable. We also consider the case in which the transition probabilities for the Markov jump process contain uncertainties. We provide a new method that estimates bounds for uncertain Markov transition probability matrix to guarantee the system stability. Numerical example verifies the computational efficiency of the proposed methods. © 2015 Elsevier B.V. All rights reserved. 1. Introduction A networked control system (NCS) is a system that is controlled over a communication network. Recently, NCSs have attracted con- siderable research interest due to emerging networked control applications. For example, NCSs are broadly used in applications including traffic monitoring, networked autonomous mobile agents, chemical plants, sensor networks, and distributed software systems in cloud computing architectures. Due to the communica- tion network, communication delays or communication losses may occur, resulting in performance degradation or even instability. Therefore, it has led various researchers to analyze NCSs associated with communication delays [ 1–8 ]. Particularly in [ 6 ], the NCS with communication delays was analyzed by adopting the switched sys- tem [ 7–11 ], which refers to the dynamical system consists of a family of subsystems and a switching logic governing switching between subsystems. In this paper, we study large-scale distributed networked con- trol system (DNCS), which denotes NCS with a large number of spatially distributed subsystems (or agents). For such large-scale systems, our primary goal is to analyze system stability when ran- dom communication delays exist. Typically, the system behavior Corresponding author. E-mail addresses: animodor@tamu.edu (K. Lee), raktim@tamu.edu (R. Bhattacharya). http://dx.doi.org/10.1016/j.sysconle.2015.08.011 0167-6911/ © 2015 Elsevier B.V. All rights reserved. with random communication delays have been widely modeled as Markov jump linear system (MJLS) [ 6 , 12–16 ] where the switching sequence is governed by a Markovian process. Since stability has been one of the major concerns, considerable effort has been on stability analysis of MJLS [ 17 , 10 , 18–20 ]. However, these results are only applicable to the systems with a small number of switching modes. Large-scale DNCSs, in which we are particularly interested, give rise to an extremely large number of switching modes. Thus, previous conditions developed for the stability analysis of MJLSs cannot be evaluated for large-scale DNCSs as they are not compu- tationally tractable . Although the literature [ 21 ] recently investi- gated the switched system that circumvents computation issues associated with a large number of switching modes, it is devel- oped for independent and identically distributed (i.i.d.) switching. We consider Markovian switching in this paper. In addition, we are also interested in large-scale DNCSs where the transition probabil- ities are inaccurately known [ 20 , 22 , 23 ]. This can happen because in practice it is difficult to accurately estimate the Markov tran- sition probability matrix that models the random communication delays. This paper provides two key contributions to analyze the sta- bility of large-scale DNCSs with random communication delays. Firstly, we guarantee the mean square stability of such systems by introducing a reduced mode model. We prove that the mean square stability for individual switched system implies a neces- sary and sufficient stability condition for the entire DNCS. This drastically reduces the number of modes necessary for analysis. " id="pdf-obj-0-5" src="pdf-obj-0-5.jpg">

Contents lists available at ScienceDirect

Systems & Control Letters

<a href=Systems & Control Letters 85 (2015) 77–83 Contents lists available at ScienceDirect Systems & Control Letters journal homepage: www.elsevier.com/locate/sysconle Stability analysis of large-scale distributed networked control systems with random communication delays: A switched system approach Kooktae Lee , Raktim Bhattacharya Department of Aerospace Engineering, Texas A&M University, College Station, TX 77843-3141, USA a r t i c l e i n f o Article history: Received 3 December 2014 Received in revised form 22 April 2015 Accepted 10 August 2015 Available online 18 October 2015 Keywords: Large-scale distributed networked control system Markov jump linear system Switched system Random communication delays a b s t r a c t In this paper, we consider the stability analysis of large-scale distributed networked control systems with random communication delays. The stability analysis is performed in the switched system framework, particularly as the Markov jump linear system. There have been considerable research on stability analysis of the Markov jump systems. However, these methods are not applicable to large-scale systems because large numbers of subsystems result in extremely large number of switching modes. To circumvent this scalability issue, we propose a new reduced mode model for stability analysis, which is computationally scalable. We also consider the case in which the transition probabilities for the Markov jump process contain uncertainties. We provide a new method that estimates bounds for uncertain Markov transition probability matrix to guarantee the system stability. Numerical example verifies the computational efficiency of the proposed methods. © 2015 Elsevier B.V. All rights reserved. 1. Introduction A networked control system (NCS) is a system that is controlled over a communication network. Recently, NCSs have attracted con- siderable research interest due to emerging networked control applications. For example, NCSs are broadly used in applications including traffic monitoring, networked autonomous mobile agents, chemical plants, sensor networks, and distributed software systems in cloud computing architectures. Due to the communica- tion network, communication delays or communication losses may occur, resulting in performance degradation or even instability. Therefore, it has led various researchers to analyze NCSs associated with communication delays [ 1–8 ]. Particularly in [ 6 ], the NCS with communication delays was analyzed by adopting the switched sys- tem [ 7–11 ], which refers to the dynamical system consists of a family of subsystems and a switching logic governing switching between subsystems. In this paper, we study large-scale distributed networked con- trol system (DNCS), which denotes NCS with a large number of spatially distributed subsystems (or agents). For such large-scale systems, our primary goal is to analyze system stability when ran- dom communication delays exist. Typically, the system behavior Corresponding author. E-mail addresses: animodor@tamu.edu (K. Lee), raktim@tamu.edu (R. Bhattacharya). http://dx.doi.org/10.1016/j.sysconle.2015.08.011 0167-6911/ © 2015 Elsevier B.V. All rights reserved. with random communication delays have been widely modeled as Markov jump linear system (MJLS) [ 6 , 12–16 ] where the switching sequence is governed by a Markovian process. Since stability has been one of the major concerns, considerable effort has been on stability analysis of MJLS [ 17 , 10 , 18–20 ]. However, these results are only applicable to the systems with a small number of switching modes. Large-scale DNCSs, in which we are particularly interested, give rise to an extremely large number of switching modes. Thus, previous conditions developed for the stability analysis of MJLSs cannot be evaluated for large-scale DNCSs as they are not compu- tationally tractable . Although the literature [ 21 ] recently investi- gated the switched system that circumvents computation issues associated with a large number of switching modes, it is devel- oped for independent and identically distributed (i.i.d.) switching. We consider Markovian switching in this paper. In addition, we are also interested in large-scale DNCSs where the transition probabil- ities are inaccurately known [ 20 , 22 , 23 ]. This can happen because in practice it is difficult to accurately estimate the Markov tran- sition probability matrix that models the random communication delays. This paper provides two key contributions to analyze the sta- bility of large-scale DNCSs with random communication delays. Firstly, we guarantee the mean square stability of such systems by introducing a reduced mode model. We prove that the mean square stability for individual switched system implies a neces- sary and sufficient stability condition for the entire DNCS. This drastically reduces the number of modes necessary for analysis. " id="pdf-obj-0-16" src="pdf-obj-0-16.jpg">

Stability analysis of large-scale distributed networked control systems with random communication delays: A switched system approach

Kooktae Lee , Raktim Bhattacharya

<a href=Systems & Control Letters 85 (2015) 77–83 Contents lists available at ScienceDirect Systems & Control Letters journal homepage: www.elsevier.com/locate/sysconle Stability analysis of large-scale distributed networked control systems with random communication delays: A switched system approach Kooktae Lee , Raktim Bhattacharya Department of Aerospace Engineering, Texas A&M University, College Station, TX 77843-3141, USA a r t i c l e i n f o Article history: Received 3 December 2014 Received in revised form 22 April 2015 Accepted 10 August 2015 Available online 18 October 2015 Keywords: Large-scale distributed networked control system Markov jump linear system Switched system Random communication delays a b s t r a c t In this paper, we consider the stability analysis of large-scale distributed networked control systems with random communication delays. The stability analysis is performed in the switched system framework, particularly as the Markov jump linear system. There have been considerable research on stability analysis of the Markov jump systems. However, these methods are not applicable to large-scale systems because large numbers of subsystems result in extremely large number of switching modes. To circumvent this scalability issue, we propose a new reduced mode model for stability analysis, which is computationally scalable. We also consider the case in which the transition probabilities for the Markov jump process contain uncertainties. We provide a new method that estimates bounds for uncertain Markov transition probability matrix to guarantee the system stability. Numerical example verifies the computational efficiency of the proposed methods. © 2015 Elsevier B.V. All rights reserved. 1. Introduction A networked control system (NCS) is a system that is controlled over a communication network. Recently, NCSs have attracted con- siderable research interest due to emerging networked control applications. For example, NCSs are broadly used in applications including traffic monitoring, networked autonomous mobile agents, chemical plants, sensor networks, and distributed software systems in cloud computing architectures. Due to the communica- tion network, communication delays or communication losses may occur, resulting in performance degradation or even instability. Therefore, it has led various researchers to analyze NCSs associated with communication delays [ 1–8 ]. Particularly in [ 6 ], the NCS with communication delays was analyzed by adopting the switched sys- tem [ 7–11 ], which refers to the dynamical system consists of a family of subsystems and a switching logic governing switching between subsystems. In this paper, we study large-scale distributed networked con- trol system (DNCS), which denotes NCS with a large number of spatially distributed subsystems (or agents). For such large-scale systems, our primary goal is to analyze system stability when ran- dom communication delays exist. Typically, the system behavior Corresponding author. E-mail addresses: animodor@tamu.edu (K. Lee), raktim@tamu.edu (R. Bhattacharya). http://dx.doi.org/10.1016/j.sysconle.2015.08.011 0167-6911/ © 2015 Elsevier B.V. All rights reserved. with random communication delays have been widely modeled as Markov jump linear system (MJLS) [ 6 , 12–16 ] where the switching sequence is governed by a Markovian process. Since stability has been one of the major concerns, considerable effort has been on stability analysis of MJLS [ 17 , 10 , 18–20 ]. However, these results are only applicable to the systems with a small number of switching modes. Large-scale DNCSs, in which we are particularly interested, give rise to an extremely large number of switching modes. Thus, previous conditions developed for the stability analysis of MJLSs cannot be evaluated for large-scale DNCSs as they are not compu- tationally tractable . Although the literature [ 21 ] recently investi- gated the switched system that circumvents computation issues associated with a large number of switching modes, it is devel- oped for independent and identically distributed (i.i.d.) switching. We consider Markovian switching in this paper. In addition, we are also interested in large-scale DNCSs where the transition probabil- ities are inaccurately known [ 20 , 22 , 23 ]. This can happen because in practice it is difficult to accurately estimate the Markov tran- sition probability matrix that models the random communication delays. This paper provides two key contributions to analyze the sta- bility of large-scale DNCSs with random communication delays. Firstly, we guarantee the mean square stability of such systems by introducing a reduced mode model. We prove that the mean square stability for individual switched system implies a neces- sary and sufficient stability condition for the entire DNCS. This drastically reduces the number of modes necessary for analysis. " id="pdf-obj-0-25" src="pdf-obj-0-25.jpg">

Department of Aerospace Engineering, Texas A&M University, College Station, TX 77843-3141, USA

a r t i c l e

i n f o

Article history:

Received 3 December 2014 Received in revised form 22 April 2015 Accepted 10 August 2015 Available online 18 October 2015

Keywords:

Large-scale distributed networked control system Markov jump linear system Switched system Random communication delays

a b s t r a c t

In this paper, we consider the stability analysis of large-scale distributed networked control systems with random communication delays. The stability analysis is performed in the switched system framework, particularly as the Markov jump linear system. There have been considerable research on stability analysis of the Markov jump systems. However, these methods are not applicable to large-scale systems because large numbers of subsystems result in extremely large number of switching modes. To circumvent this

scalability issue, we propose a new reduced mode model for stability analysis, which is computationally

scalable. We also consider the case in which the transition probabilities for the Markov jump process contain uncertainties. We provide a new method that estimates bounds for uncertain Markov transition probability matrix to guarantee the system stability. Numerical example verifies the computational efficiency of the proposed methods.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

A networked control system (NCS) is a system that is controlled over a communication network. Recently, NCSs have attracted con- siderable research interest due to emerging networked control applications. For example, NCSs are broadly used in applications including traffic monitoring, networked autonomous mobile agents, chemical plants, sensor networks, and distributed software systems in cloud computing architectures. Due to the communica- tion network, communication delays or communication losses may occur, resulting in performance degradation or even instability. Therefore, it has led various researchers to analyze NCSs associated with communication delays [1–8]. Particularly in [6], the NCS with communication delays was analyzed by adopting the switched sys- tem [7–11], which refers to the dynamical system consists of a family of subsystems and a switching logic governing switching between subsystems. In this paper, we study large-scale distributed networked con- trol system (DNCS), which denotes NCS with a large number of spatially distributed subsystems (or agents). For such large-scale systems, our primary goal is to analyze system stability when ran- dom communication delays exist. Typically, the system behavior

Corresponding author. E-mail addresses: animodor@tamu.edu (K. Lee), raktim@tamu.edu (R. Bhattacharya).

0167-6911/© 2015 Elsevier B.V. All rights reserved.

with random communication delays have been widely modeled as Markov jump linear system (MJLS) [6,12–16] where the switching sequence is governed by a Markovian process. Since stability has been one of the major concerns, considerable effort has been on stability analysis of MJLS [17,10,18–20]. However, these results are only applicable to the systems with a small number of switching modes. Large-scale DNCSs, in which we are particularly interested, give rise to an extremely large number of switching modes. Thus, previous conditions developed for the stability analysis of MJLSs cannot be evaluated for large-scale DNCSs as they are not compu- tationally tractable. Although the literature [21] recently investi- gated the switched system that circumvents computation issues associated with a large number of switching modes, it is devel- oped for independent and identically distributed (i.i.d.) switching. We consider Markovian switching in this paper. In addition, we are also interested in large-scale DNCSs where the transition probabil- ities are inaccurately known [20,22,23]. This can happen because in practice it is difficult to accurately estimate the Markov tran- sition probability matrix that models the random communication delays. This paper provides two key contributions to analyze the sta- bility of large-scale DNCSs with random communication delays. Firstly, we guarantee the mean square stability of such systems by introducing a reduced mode model. We prove that the mean square stability for individual switched system implies a neces- sary and sufficient stability condition for the entire DNCS. This drastically reduces the number of modes necessary for analysis.

  • 78 K. Lee, R. Bhattacharya / Systems & Control Letters 85 (2015) 77–83

Secondly, we present a new method to estimate the bound for un- certain Markov transition probability matrix, in order to guarantee the system stability. These results enable us to analyze large-scale systems in a computationally tractable manner. Rest of this paper is organized as follows. We introduce the problem for the large-scale DNCS in Section 2. Section 3 presents the switched system framework for the stability analysis with communication delays. In Section 4, we propose the reduced mode model to efficiently analyze stability. Section 5 quantifies the stability region and bounds for uncertain Markov transition probability matrix. This is followed by the application of the proposed method to an example system in Section 6, and we conclude the paper with Section 7.

Notation. The set of real numbers is denoted by R. The symbols ∥·∥ and ∥ · ∥ stand for the Euclidean and infinity norm, respectively. The symbol #(·) denotes the cardinality—the total number of elements in the given set. In addition, the symbols tr(·), ρ(·), , and diag(·) represent trace operator, spectral radius, Kronecker product, and block diagonal matrix operator, respectively.

2. Problem formulation

  • 2.1. Distributed networked control system with no delays

agent i with communication delays can be expressed as:

x i (k + 1) = A ij x j (k ),

jN i

i = 1, 2,

. . .

, N,

(3)

where k k τ . Note that we have no communication delays when i = j because there is no communication in this case. The random communication delay, represented by the term k , forms a stochastic process. To analyze the stability of the DNCS, we define an augmented state X(k) as X(k) [x(k) , x(k

1) ,

. . .

, x(k τ d ) ] R Nnq×Nnq , where q τ d + 1. Then, the

dynamics for the entire system is given by

X(k + 1) = W(k)X(k),

(4)

where W(k)

 

A ˜ 1 (k)


I
0

0

I

·

·

·

·

·

·

·

·

·

0

0

0

0

A ˜ 2 (k)

A ˜ q1 (k)

A ˜ q (k)

R Nnq×Nnq

,

.

.

.

.

.

.

0

.

.

.

·

·

·

.

.

.

I

.

.

.

0

 

0

the matrix I denotes an identity matrix with proper dimensions,

and the time-varying matrices

˜

A j (k) R Nn×Nn , j =

1, 2,

. . .

, q,

model the randomness in the communication delays between

neighboring agents.

  • 3. Switched system approach

Consider a discrete-time dynamics of each agent in the DNCS, given by:

Without loss of generality, the dynamics of the large-scale DNCS with communication delays in (4) can be transformed into

x i (k + 1) = A ij x j (k),

 

i = 1, 2,

 

, N,

(1)

a switched system framework as follows:

 

jN i

 

x(k + 1) = W σ(k) x(k),

σ(k) ∈ {1, 2,

, m},

(5)

where k is a discrete-time index, N is the total number of agents (subsystems), x i R n is a state for the ith agent, N i is a set of neighbors for x i including the agent x i itself, and A ij R n×n is a time-invariant system matrix that represents the linear intercon- nections between agents. Note that we have A ij = 0 if there is no interconnection between the agents i and j. To represent the entire systems dynamics, we define the state

where W σ(k) is the time-invariant matrix, representing commu- nication delays in agents, {σ(k)} is the switching sequence, and m is the total number of switching modes. When the switch- ing sequence {σ(k)} is stochastic, (5) is referred to as a stochas- tic switched linear system or a stochastic jump linear system, according to the literature [7]. For the stochastic switched lin- ear system, the switching sequence {σ(k)} is governed by the

x(k) R Nn×Nn as x(k) [x 1 (k) , x 2 (k) ,

, x N (k) ] . Then, the

mode-occupation switching probability π(k) = [π 1 (k), π 2 (k),

system dynamics of the DNCS is given as

, π m (k)], where π i is a fraction number, satisfying

  • m = 1

i=1 π i

x(k + 1)

= Ax(k),

 

(2)

and 0 π i 1, i. In this case, each π i denotes the modal

framework has been widely employed [12–16]. Thus, we make the

with the following definition for the matrix A R Nn×Nn

probability corresponding to each mode dynamics W i . In order to properly describe the behavior of random communication delays, it is necessary to adopt a certain switching logic, which is used to

 

   

A

A

A

11

A

12

A

13

·

·

·

A

1N

 

21

A

A

22

A

A

23

·

·

·

A

A

2N

update the switching probability π(k). For this purpose, the MJLS

A

31

32

33

·

·

·

3N

   

,

following assumption in our analysis.

. . .

 

. . .

 

. . .

.

.

 
 

.

.

. .

A N1

A ij =

0,

A ij ,

A N2

A N3

·

·

·

A NN

if no connection between the agents i and j,

Assumption: Consider the stochastic jump linear system (5) with the switching probability π(k) = [π 1 (k), π 2 (k), π m (k)]. Then, π(k) is updated by the Markovian process given

,

otherwise.

by π(k+1) = π(k)P, where P R m×m is the Markov transition probability matrix.

For the discrete-time system in (2), it is well known that the system is stable if and only if the condition ρ(A) < 1 is satisfied. Throughout the paper we assume that the system without commu- nication delays, defined in (2), is stable. Then, we address stability in the presence of random communication delays. We remind the reader that N is very large.

  • 2.2. DNCS with communication delays

Often, network communication between agents encounter time delays or packet losses while sending and receiving data. We

denote the symbol τ as random communication delays and assume

that τ has a discrete value bounded by 0

τ τ d < , where

τ d is a finite-valued maximum delay. Then, the dynamics for the

Since the MJLS is a family of the stochastic switched linear system, various stability notions can be defined [10]. In this paper, we will consider the mean square stability condition, defined below.

Definition 3.1 (Definition 1.1 in [11]). The MJLS is said to be mean square stable if for any initial condition x 0 and arbitrary initial probability distribution π(0), lim k E x(k, x 0 )2 = 0.

The total number of switching modes m depends on the size q and N. Since the communication delays take place independently while receiving and sending the data for each agent, m is calculated by counting all possible scenarios to distribute every matrices A ij R n×n for i ̸= j in the block matrix A R Nn×Nn given in (2), into

K. Lee, R. Bhattacharya / Systems & Control Letters 85 (2015) 77–83

79

˜

each A j (k) R Nn×Nn , j = 1, 2,

. . .

, q, given in (4), which results in

m = q N(N1) . Note that the large-scale DNCS that has considerably

large N results in extremely large m, which makes current analysis

tools useless as they are computationally intractable.

Before we further proceed, we introduce the following propo-

sition that was developed for the stability analysis of the MJLS.

Proposition 3.1 (Theorem 1 in [18]). The MJLS with the Markov

transition probability matrix P is mean square stable if and only if

ρ P I diag(W j W j ) < 1,

(6)

where I is an identity matrix with a proper dimension,

diag(W j W j )

where P i R m i ×m i is the transition probability matrix for the

reduced mode MJLS given in (7), I is an identity matrix with a proper

dimension, N is the total number of the agents in the system, m i =

q nˆ i (nˆ i 1) is the total mode numbers for the reduce mode MJLS, and

diag( W

j

ˆ

i

ˆ

i

W

j

)

(

ˆ

W

i

1

0

.

.

.

0

0

ˆ

W

i

1 )

0

(

ˆ

i

W

2

ˆ

i

2 )

W

0

0

0

0

.

.

.

.

.

.

·

·

·

·

·

·

( W ˆ

i

m i -1

0

m i -1 )

W

ˆ

i

0

0

.

.

.

0

(

m i m i )

W

W

ˆ

i

ˆ

i

.

Proof. Let the matrix Q i (k) be of the form Q i (k) E[xˆ i (k)xˆ i (k) ].

Then, Q i (k) is alternatively obtained by the following equation:

  (W 1 ⊗ W 1 ) 0 0 · · · 0 m i
(W 1 ⊗ W 1 )
0
0
·
·
·
0
m
i
i
Q i (k) = 
i
(k), where Q
(k)
E  xˆ i (k)xˆ i (k) ⊤ |σ i (k) = s 
0
(W 2 ⊗ W 2 )
0
·
·
·
s=1 Q s
s
0
 
i
i
i
.
.
π
(k), and π (k) Pr  σ i (k) = s  . Then, Q (k) satisfies
.
,
s
s
s
. . .
.
. .
      
    
m
i
0
0
0
(W m-1 ⊗ W m-1 )
i
Q
(k) =
E[xˆ i (k)xˆ i (k) ⊤ | σ i (k) = s, σ i (k − 1) = r]
0
0
.
.
.
0
(W m ⊗ W m )
s
r=1
and m is the total number of the switching modes.
i
Pr(σ i (k − 1) = r | σ i (k) = s)π (k)
s
m
i
m
For the given set of matrices {W σ(k) } σ(k)=1 and the transition
=
E[xˆ i (k)xˆ i (k) ⊤ | σ i (k) = s, σ i (k − 1) = r]
probability matrix P, one can always compute the spectral radius
r=1
given in (6), and hence guarantee the system stability.
i
Pr(σ  i (k) = s |  σ i (k − 1) = r)  π (k − 1)
r
Unfortunately, this condition is not applicable to large-scale
DNCSs due to enormously large m. For example, even with q =
p rs i
2 and N = 100, we have m = 2 100×99 . It is not practically
m
i
i
i
possible to compute the spectral radius in (6) for such problems.
=
p
rs E[xˆ i (k)xˆ i (k) ⊤ | σ i (k) = s, σ i (k − 1) = r]π (k − 1)
r
To circumvent this scalability issue, we present a new analysis
r=1
m
approach for such large-scale DNCSs in the next section.
i
i
ˆ
i
ˆ
i
=
p
W
σ i (k−1) xˆ i (k − 1)xˆ i (k − 1) ⊤
W
rs E[
σ i (k−1) |
r=1
4. Stability with reduced mode dynamics
i
σ i (k − 1) = r]π
(k − 1)
r
m
i
In this section, we define a new augmented state to reduce the
i
ˆ
i
i
ˆ
i
=
p
W
(k − 1)
W
rs
r
E[xˆ i (k − 1)xˆ i (k − 1) ⊤ | σ i (k − 1) = r]π
 
r
r
mode numbers as follows:
r=1
i
=Q r (k−1)
xˆ i (k) [x˜ i (k) ⊤ , x˜ i (k − 1) ⊤ , .
. .
, x˜ i (k −
τ d ) ⊤ ] ⊤ ∈ R nˆ i nq ,
m
i
i
ˆ
i
i
ˆ
i
=
p
W Q
(k − 1)
W
.
rs
r
r
r
where nˆ i #(N i ), N i stands for the set of neighbors to x i including
r=1
itself, x˜ i (k) [x i (k) ⊤ , x j (k) ⊤ ] ⊤ ∈ R nˆ i n , and x j (k) ∈ R n , j ∈ N i ,
In the second equality of above equation, p rs i denotes the mode
denotes all states that are neighbor to x i (k) ∈ R n .
transition probability from r to s in the Markov transition
Then, we can construct a switched linear system framework
probability matrix P i .
similarly to (5) as follows:
Taking the vectorization in above equation results in
ˆ
i
xˆ i (k + 1) =
W
σ i (k) xˆ i (k),
σ i (k) ∈ {1, 2,
. . .
, m i },
(7)
m
i
i
ˆ
i
i
ˆ
i
vec  Q
(k)  = vec 
p i W Q
(k − 1)
W
ˆ
s
rs
r
r
r
A ˆ 1 (k)
A ˆ 2 (k)
·
·
·
(k)
A ˆ q (k)
A q−1
r=1
I
0
·
·
·
0
0
ˆ
i
0
I
·
·
·
0
0
m
R nˆ i nq×nˆ i nq
i
where W
σ i (k)
ˆ
i
i
ˆ
i
.
.
.
.
.
=
p rs i vec  W Q
(k − 1)
W
.
.
.
.
.
r
r
r
.
.
.
.
.
r=1
0
0
·
·
·
I
0
m
i
ˆ
with the time-varying matrix
A j (k) ∈ R nˆ i n×nˆ i n , j = 1, 2,
, q.
In
ˆ
i
ˆ
i
i
=
p i
W
W )vec(Q
(k − 1)).
rs (
r
r
r
this case, the total number of the switching modes for (7) is given
r=1
by m i = q nˆ i (nˆ i −1)
.
By implementing the reduced mode model given in (7), we
In the last equality, we used the property that vec(ABC) = (C ⊤ ⊗
provide a computationally efficient condition for stability analysis
A)vec(B).
of the original DNCS in the following theorem.
i
Now we define a new variable y i (·) (k) vec  Q (·) (k)  , which
leads to
Theorem 4.1. Consider the large-scale DNCS (5) with Markovian
m
i
communication delays associated with the transition probability
i
ˆ
i
ˆ
i
y
s (k) =
p i
W
W
)y r i (k − 1).
rs (
r
r
matrix P. The necessary and sufficient condition for the mean square
r=1
stability of this system is given by
By stacking y (·) i (k) from 1 up to m i , with a new definition for the
ˆ
i
ˆ
i
ρ  (P i ⊤ ⊗ I)diag(
W
W
)  < 1, ∀i = 1, 2,
. . .
, N,
(8)
augmented state yˆ i (k)
[y i
1 (k) ⊤ y 2 i (k) ⊤
. . .
y i
m i (k) ⊤ ] ⊤ , we have
j
j
  • 80 K. Lee, R. Bhattacharya / Systems & Control Letters 85 (2015) 77–83

the following recursion equation:

yˆ i (k)

 

=    

p

i

11

(

ˆ

W

i

1

p

i

12

(

ˆ

W

i

1

.

. .

p i

1m i (

ˆ

W

i

1

ˆ

W

i

1

)

ˆ

W

i

1

)

ˆ

W

i

1 )

p

i

21

(

ˆ

i

W

2

p

i

22

(

ˆ

i

W

2

ˆ

i

2 )

W

ˆ

i

2 )

W

. . .

p

i

2m i (

ˆ

i

W

2

ˆ

i

2 )

W

.

.

.

.

.

.

.

.

.

.

.

.

p i

m i 1 (

m i m i )

W

W

ˆ

i

ˆ

i

p i

m i 2 (

m i m i )

W

W

ˆ

i

ˆ

i

. . .

p i

m i m i (

m i m i )

W

W

ˆ

i

ˆ

i

 

y i

1 (k 1)

y i

2 (k 1)

  

  

   

y i

.

. .

m i (k 1)

 

   

.



 



=(P i I)diag( W

ˆ

i

j

i

ˆ

W

j

)

=yˆ i (k1)

From the above equation, it is clear that ρ (P i I)diag( W

ˆ

i

j

W )

ˆ

i

j

<

1 implies lim k yˆ i (k)

=

0, and hence this leads to

lim k Q i (k) = 0 ⇐⇒ lim k tr Q i (k) = 0 ⇐⇒ lim k E

xˆ i (k)2 = 0, which is the sufficient mean square stabil-

ity condition for xˆ i (k). On the other hand, if we have ρ (P i

I)diag( W

j

ˆ

i

W ) >

ˆ

i

j

1, then yˆ i (k) will diverge, resulting in ne-

cessity for the mean square stability of xˆ i (k). Hence, the spectral

radius being less than one is the necessary and sufficient mean

square stability condition for the state xˆ i (k). Further, we have

lim k E xˆ i (k)2 = 0, i = 1, 2,

. . .

, N ⇐⇒ lim k E

x(k)2 = 0, where x(k) is the state for the DNCS defined in (5).

This concludes the proof.

Remark 4.1. Theorem 4.1 provides an efficient way to analyze

the stability of large-scale DNCSs. The key idea stems from

the hypothesis that the stability of each subsystem obtained

by decomposing the original system will provide the stability

of the entire system. Without any relaxation or conservatism,

Theorem 4.1 provides the necessary and sufficient condition for

stability, which is equivalent to (6) developed for the mean square

stability of the MJLS. Compared to the total number of modes in

the full state model (5), which is q N(N1) , the reduced mode model

N

(7) has total i=1

q nˆ i (nˆ i 1) modes. Consequently, the growth of

mode numbers in full state model is exponential with respect to

N 2 , whereas that in reduced mode model is linear with regard to

N. Therefore, Theorem 4.1 can be checked in a computationally

efficient manner.

5. Stability region and stability bound for uncertain Markov

transition probability matrix

The Markov transition probability matrix can be obtained from

data of communication delays. In general, this Markov transition

probability matrix obtained from statistics can be interpreted as

representation of random communication delays in an average

manner. Thus, one cannot estimate the exact transition probability

in practice, which leads to uncertainty in the Markov transition

probability matrix. In this section, we aim at developing a tool

to measure the stability bound for uncertain Markov transition

probability matrix, in order to guarantee the stability of large-scale

DNCSs with uncertain Markovian communication delays.

As explained in Section 3, the dimension of the Markov

transition probability matrix for the full state model is given by

P R m×m , where the number of switching modes is m = q N(N1) .

For fairly large N, it is practically intractable to handle such a

large-scale matrix. For example, even with q = 2 and N =

100, the mode number is given by m = 2 100×99 , resulting in

P R 2 100×99 ×2 100×99 . However, the reduced mode model developed

in Section 3 renders computation tractable because it results in

huge decrease in the mode numbers as well as the dimension of

each Markov transition probability matrix, by decomposing the

original system into N numbers of reduced mode systems. As a

consequence, the reduced mode model yields N numbers of new

Markov transition probability matrix P i , i = 1, 2,

has small size.

. . .

, N, which

Here we assume that the new Markov transition probability

matrix associated with the each reduced mode is given by the

following form: P i =

¯

P i + P i , i

=

1, 2,

. . .

¯

, N, where P i is the

nominal value and P i is the uncertainty in the Markov transition

probability matrix for ith subsystem. Due to the variation in P i ,

the system stability may change and hence we want to estimate the

bound for P i , to guarantee the system stability. Here we assume

that P i has the following structure:

P i

p

i

11

p

i

21

.

.

.

p

i

m i 1

p

i

12

p

i

22

.

.

.

p

i

m i 2

·

·

·

·

·

·

.

.

.

·

·

·

p

p

i

1m

i

i

2m

i

.

.

.

p i

m i m i

, R m i ×m i

s.t.

m

i

s=1

p i

rs = 0, r = 1, 2,

. . .

,

m i .

(9)

Since we have a constraint such that the row sum has to be zero

for P i in above equation, we aim to find the feasible maximum

bound for each row, ε r , satisfying the inequality | p i

i

rs | ≤ ε r , r,

i

i

in order to guarantee the system stability. Then, each ε r for r =

1, 2,

. . .

, m i can be obtained by the following two steps.

Step 1: Solve Linear Programming (LP)

 

maximize

1

z

 

(for upper bound)

 

(10)

or minimize

1

z

(for lower bound)

 
 

subject to

A|z| < b s ,

s = 1, 2,

, m i

(11)

 

lb s z s ub s , s = 1, 2,

 

, m i

(12)

where

 

z s [ p i

1s , p i

2s ,

z [z

i ]

, p i

m i s ]

,

1

, z

2

,

.

.

.

, z

m

,

A α 1 , α 2 ,

j = 1, 2,

, α m i , with α j

ˆ

W

, m i ,

 

ˆ

i

j

W

j

  • i ,

 

m

i

b s 1

α r p¯ i

rs ,

r=1

lb s [−p¯ i 1s , p¯ 2s i ,

, p¯ i

m i s ]

,

ub s [1 p¯ i

1s , 1 p¯ 2s

i

,

.

.

.

, 1

m i s ]

p¯ i

.

The inequality constraint (11) in the LP problem guarantees the

mean square stability by Lemma 5.1 and Theorem 5.1. The terms lb s

and ub s in (12) are the lower and upper bounds for z s , respectively,

according to 0 p i

rs = (p¯ i

rs + p i

rs ) 1.

Step 2: Obtain Feasible Solution with Hyperplane Constraint

 
 

We can compute the feasible maximum bound for p i

rs

as

follows.

 

ε

i

r = min min(|ε

r,lb |), min(|ε r,ub |) ,

i

i

r = 1, 2,

, m i .

(13)

where ε

r,lb [(p i

i

r1 )

lb ,(p i

r2 )

lb ,

,(p i

rm ) lb ]

,

ε

r,ub [( p i

i

r1 )

ub ,(p i

r2 )

ub ,

,(p i

rm )

ub ] , and (p i

lb ,(p i

rs )

rs ) ub

denote optimal lower and upper bounds for p rs i , obtained from

the LP, respectively.

 

Since upper or lower bounds are solved by maximizing or

minimizing the objective function, ( p rs i ) has different values for

upper and lower bounds. Fig. 1 shows the geometry of stability

region analysis for uncertain transition probability matrix. The

K. Lee, R. Bhattacharya / Systems & Control Letters 85 (2015) 77–83

81

K. Lee, R. Bhattacharya / Systems & Control Letters 85 (2015) 77–83 81 Fig. 1. The

Fig. 1. The geometry of the Stability Region Analysis for the uncertain Markov transition probability matrix when m i = 2. Each region is described in the figure.

region S 1 stands for the bounds that come from p¯ i

rs

p i

rs

1 p¯ i

rs . S 2 can be obtained from inequality constraint (11). The

region S 3 denotes the solution from the LP, and S is the feasible

maximum bound with a stability guarantee. Note that P i satisfies

  • m s=1 p i

i

rs = 0, r and hence, feasible solutions should lie on the

hyperplane, satisfying p i

r1 + p i

r2 +· · ·+ p i

rm

i

= 0, r. Therefore,

we can compute the feasible maximum bound from (13) for each

row r.

Now we prove that the inequality constraint (11) guarantees

the system stability. For this purpose, Lemma 5.1 will be used.

Lemma 5.1. Consider a block matrix X defined by

X =

X 11

X 21

.

.

.

X m1

X 12

X 22

.

.

.

X m2

·

·

·

·

·

·

.

.

.

X 1m

X 2m

.

.

.

· · · X mm

,

where matrix X ij R n×n . Then, we have ρ (X) < 1, if

< 1, i = 1, 2,

. . .

, m.

  • m

j=1

X ij

Proof. For the block matrix X given above, the following inequality

m

condition X max i j=1 X ij holds. Also, it is well known

that ρ(X) ≤ ∥Xp for any choice of p.

Therefore, we conclude that

m

j=1 X ij < 1, i = 1, 2,

. . .

, m ρ(X) ≤ ∥X < 1.

Theorem 5.1. Consider the MJLS (5) for the large-scale DNCS with

communication delays. Then, (5) is mean square stable if

m

i

r=1

α r | p rs

i | < β s ,

s = 1, 2,

. . .

, m i ,

i = 1, 2,

. . .

, N

where α r = ∥ W

r

ˆ

i

satisfied.

ˆ

i

W

r

and β s = 1 m

i

r=1 p¯ rs

i

ˆ

i

W

r

W , is

r

ˆ

i

Proof. If the Markov transition probability matrix for the system

¯

in (7) has the uncertainty denoted by P i = P i + P i , then the term

ρ (P i

I)diag(

ˆ

i

W

j

W ) in (8) can be expressed as

ˆ

i

j

ρ (P i

I)diag(

ˆ

i

W

j

ˆ

i

W

j

)

= ρ (

ˆ

P i + P i ) I diag( W

¯

i

j

ˆ

i

W

j

)

=

ρ   (

¯

P

i I) + (P i I) diag(

ˆ

i

W

j

ˆ

i

W

j

)

= ρ (

¯

P

i

I)diag( W

j

ˆ

i

ˆ

i

W

j

) + (P i I)diag(

ˆ

i

W

j

ˆ

i

W

j

)

≤ ∥(

¯

P

i

I)diag( W

j

ˆ

i

ˆ

i

W

j

) + (P i I)diag(

ˆ

i

W

j

ˆ

i

W

j

)

≤ ∥(

¯

P

  • i

I)diag( W

j

ˆ

i

ˆ

i

W

j

)

+ ∥(P i I)diag(

ˆ

i

W

j

W ) .

j

ˆ

i

(14)

In the first inequality, we used the fact that ρ(·) ≤ ∥ · ∥ and

the sub-multiplicative property was applied in the last inequality.

The block matrix structure for each term of the last inequality

is alternatively expressed as the equation in Box I, where γ j =

(

ˆ

i

W

j

W ), j = 1, 2,

j

ˆ

i

. . .

, m i , and similarly,

(P i I)diag( W

ˆ

i

j

ˆ

i

W

j

)

=

γ 1 p 11 i

γ 1 p 12 i

.

.

.

γ 1 p 1m i i

γ 2 p i

21

γ 2 p i

22

.

.

.

γ 2 p i

2m i

γ m i p i

m i 1

γ m i p i

m i 2

.

.

.

·

·

·

·

·

·

.

.

.

· · · γ m i p i

m i m i

.

By applying the result in Lemma 5.1 into (14), it is guaranteed

that ρ (P i I)diag( W

ˆ

i

j

W ) < 1, if the following condition

ˆ

i

j

m

i

r=1

α r | p i

rs | +

m

i

r=1

i

α r p¯ rs

< 1,

s = 1, 2,

. . .

, m i ,

where α r W

r

ˆ

i

W , is satisfied.

r

ˆ

i

Therefore, (5) is mean square stable by Theorem 4.1 if it is

guaranteed that

m

i

r=1

α r | p rs

i | < β s ,

s = 1, 2,

i = 1, 2,

. . .

, m i ,

. . .

, N,

where β s 1 m

i

r=1 α r p¯ rs

i

.

  • 6. Numerical example We consider the DNCS that has random communication delays,

where the total number of agents is denoted by N = 100. The

state for each agent is updated by the following scalar discrete-

time dynamics:

x i (k +

1) =

a ij x j (k ),

jN i

i = 1, 2,

. . .

, N,

where a ij 0.7,

0,

j = i,

0.14, j

∈ {i 1, i + 1},

otherwise.

(15)

When there is no communication delay (i.e., k = k), the dynamics

for the entire DNCS is obtained by (2) as follows:

X(k + 1) = AX(k),

  

A =   

 

0.7

0.14

0

.

. .

0

0

0.14

0.7

0.14

·

·

0

0

0.14

0.7

.

.

.

·

·

·

·

0

0

0.14

.

.

.

0.14

0

·

·

·

·

·

·

·

.

.

.

0.7

0.14

0

0

0

. . .

0.14

0.7

       

R N×N

.

In this case, the stability is computed by ρ(A) = 0.9799 < 1,

which assures that the given DNCS without communication delays

is stable. However, communications (e.g., wireless network) may

be delayed in practice and thus, above stability result is no longer

guaranteed. Therefore, it is required to test the system stability in

the existence of random communication delays.

6.1.

Stability analysis with random communication delays

We assume that the communication delay τ is bounded by

0 τ

τ d

=

1, i.e., k

= {k, k 1}, i =

1, 2,

. . .

, N, which

  • 82 K. Lee, R. Bhattacharya / Systems & Control Letters 85 (2015) 77–83

             i i i
 
 
 
i
i
i
i
i
·
·
·
ˆ
ˆ
·
·
·
W
W
0
0
11 I
21 I
m i 1 I
1 ⊗
1
i
i
i
i
i
·
·
·
ˆ
ˆ
0
W
W
0
12 I
22 I
m i 2 I
2 ⊗
2
¯
ˆ
i
ˆ
i
(
i
P
⊗ I)diag( W
W
) 
=
.
.
.
.
.
j
j
 ∞
 
.
.
.
 
.
.
 
.
.
·
·
·
.
.
.
i
i
p¯ i
p¯ i
·
·
·
p¯ i
ˆ