X. Ma et al.: Predefined-Time Consensus of Nonlinear Multi-Agent Input Delay/Dynamic Event-Triggered
1) Different from the fixed-time consensus results
[21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31],
[32], [33], [34], the prescribed convergence time of the sys-
tem is independent of the controller parameters, and the
setting is simple and less conservative.
2) Compared with the static event-triggered consensus
results [6], [7], [8], [15], [16], [17], [18], [19], [20], [27], [28],
[29], [30], [32], [33], [34], [40], we introduce a dynamic
event-triggered factor to reduce the number of controller
triggers and resource consumption.
3) Different from the predefined-time consensus
results [35], [36], [37], [38], we consider the case of input
delay, event-triggered and switching topology.
This paper is structured as follows. Section II introduces
preliminaries and problem formulation. Section III consid-
ers the predefined-time consensus of nonlinear uncertainty
MASs with input delay. Section IV considers dynamic event-
triggered predefined-time consensus. Section V is the simu-
lation results and analysis. Section VI concludes this article.
II. PRELIMINARIES AND PROBLEM FORMULATION
A. GRAPH THEORY
Consider a MASs with M agents whose communication
topology is represented by an undirected graph G =
(V , ε, A), where V = {v
1
, v
2
, · · · , v
m
} is the set of nodes,
ε ⊆ V × V represents the set of edges, A = [a
ij
]
m×m
is an adjacency matrix. Each node v
i
represents an agent i,
and the existence of an edge set (v
i
, v
j
) ∈ ε indicates that
agents j and i can communicate with each other. For an
adjacency matrix A = [a
ij
]
m×m
of an undirected graph G,
when (v
j
, v
i
) ∈ ε, a
ij
= 1, otherwise a
ij
= 0. The Laplacian
matrix L of an undirected graph G is defined as L = D − A,
where D = diag{d
1
, d
2
, · · · , d
m
} is a diagonal matrix, d
j
=
P
m
i=1,i=j
a
ji
, j ∈ {1, 2, · · · , m}. An undirected graph G is said
to be connected if there is at least one path between any two
nodes.
B. DEFINITION AND LEMMAS
Consider the following system
˙x(t) = f (x(t), t)x(0) = x
0
(1)
where x = [x
1
, x
2
, · · · , x
m
]
T
∈ R
m
, f (x(t), t) : R
m
× R
+
→
R
m
is a nonlinear function. Let the origin be an equilibrium
point of system (1).
Definition 1 [41]: If the origin of system (1) is globally
uniformly finite-time stable and the convergence time param-
eter T : R
m
→ R
+
is globally bounded, then the origin
is called the equilibrium point of global fixed-time conver-
gence, i.e. there exists a finite constant T
max
∈ R
+
such that
for all t ≥ T and x
0
∈ R
m
satisfying T
s
< T
max
, x(t) = 0,
then the origin of system (1) is pre-determined time
convergent.
Lemma 1 [42]: If there exists a positive definite function
V (x) : R
m
→ R, ∀x ∈ R
m
, and V (x) = 0 ⇔ x = 0 such that
the following holds
˙
V (x) ≤ −
π
αT
(V (x)
1−
α
2
+ V (x)
1+
α
2
) (2)
where constant T > 0, 0 < α < 1, then the origin of the
system is stable for the predefined time T.
Lemma 2 [43]: A connected undirected graph G whose
Laplacian matrix L is positive semidefinite with eigenvalues
satisfying
0 = λ
1
(L) < λ
2
(L) ≤ · · · ≤ λ
m
(L)
λ
2
(L) = min
∥
x
∥
=0,
m
P
i=1
x
i
=0
x
T
Lx
∥
x
∥
2
(3)
For x =
[
x
1
, x
2
, · · · , x
m
]
T
∈ R
n
, there is
x
T
Lx =
1
2
m
X
i=1
m
X
j=1
a
ij
(x
i
− x
j
)
2
(4)
Therefore, if 1
T
x = 0, that is
m
P
i=1
x
i
= 0, then λ
2
(L)x
T
x ≤
x
T
Lx ≤ λ
m
(L)x
T
x.
Lemma 3 [44]: If |y| denotes the absolute value of the real
number y, then
d
dy
|
y
|
α+1
= (α + 1)sig(y)
α
d
dy
sig(y)
α+1
= (α + 1)
|
y
|
α
where sig(y)
α
= sign(y)
|
y
|
α
.
Lemma 4 [45]: For a real number ε
1
, ε
2
, · · · , ε
m
∈ R
+
,
one has
m
1−p
m
X
i=1
ε
i
!
p
≥
m
X
i=1
ε
p
i
≥
m
X
i=1
ε
i
!
p
, 0 < 0 ≤ 1
m
X
i=1
ε
q
i
≥ m
1−q
m
X
i=1
ε
i
!
q
, q > 1
C. PROBLEM DESCRIPTION
Consider nonlinear MASs with input delay disturbance
˙x
i
(t) = u
i
(t − τ ) + f (x
i
(t), t) + d
i
(x
i
(t), t)
x(0) = x
0
(5)
where x
i
∈ R
n
and u
i
∈ R
n
, i = 1, 2, · · · , m are state and the
control input of agent i, τ is the known input delay, f (x(t), t) :
R
m
× R
+
→ R
m
is nonlinear uncertainty, d
i
(x
i
(t), t) ∈ R
n
can represent unknown disturbance and noise, etc..
Definition 2: For the input delay nonlinear MASs (5) with
disturbance, there exists a predefined time constant T such
that when t ≥ T , the system has
(
lim
t→T
x
i
(t) − x
j
(t)
= 0
x
i
(t) = x
j
(t)
(6)
Then the system can achieve predefined time consistency.
VOLUME 11, 2023 29885