Bounds on Delay Margin for Consensus of General Second-Order
Multi-Agent Systems
Rui Tian, Dan Ma and Jie Chen
Abstract— This paper studies the delay margin and its
bounds for second-order multi-agent systems to achieve robust
consensus with respect to uncertain delays varying within a
range. The issue under investigation dwells on the question:
What is the largest delay range within which a distributed
control protocol is able to achieve and maintain the consensus?
We consider second-order agents with real unstable poles com-
municating over an undirected network topology, and derive
bounds on the delay margin. The results show that for a strictly
unstable second-order multi-agent system, its consensuability
robustness depends on the pole locations of the agents, as well
as on the eigen-ratio of the network graph.
I. INTRODUCTION
Multi-agent systems (MASs), for their widely recognized
applications in, e.g., coordination of sensor networks and
multi-robotic systems, formation control of unmanned air
vehicles (UAVs), and in distributed computation [1-3], have
received considerable attention in the recent years. In these
applications, a groups of agents are to cooperate to accom-
plish a common task. A central problem in the study of
MASs is that of achieving consensus, which amounts to
designing a suitable feedback protocol so that all the agents
in a network converge to a common state through local
interactions, whereas the common state is typically defined
as the average state of the agents, or the maximal/minimal
state of the agents, termed average-consensus and max/min-
consensus, respectively. By its intrinsic nature the agents
in a MAS configuration must exchange information over
a communication network, which invariably, is prone to
transmission delays, due to, for example, communication
congestions and transmission bandwidth. It is known that
time delay will generally degrade a system’s performance,
and this remains so, if not more acute, in a MAS. As
such, communication delays in MASs must be accounted
for. Over the past decade, MAS consensus problems with
delayed communication topology have been well studied;
see, e.g., [3, 4, 6, 8, 9-14, 16-20]. Various agent models
have been considered, including single-integrator agents [4],
double-integrator agents [6, 9, 10], high-order agents [16-
18], and further, general nonlinear agents [3]. In [4], upper
This research was supported in part by NSFC under Grants 61603079,
61773098, and in part by the Hong Kong RGC under Project CityU
11201514, CityU 111613.
Rui Tian is with College of Information Science and Engineering, North-
eastern University, Shenyang, 110819, China 364368795@qq.com
Dan Ma is with College of Information Science and
Engineering, Northeastern University, Shenyang, 110819, China
madan@mail.neu.edu.cn
Jie Chen is with Department of Electronic Engineering, City University
of Hong Kong, Hong Kong, China jichen@cityu.edu.hk
bounds on homogenous delays were obtained, which provide
a range of delay that a MAS of single-integrator agents can
achieve consensus by a constant feedback protocol over a
fixed undirected network topology. Other similar results can
be found in, e.g., [12-14], which also seek to ascertain delay
ranges to insure consensus of first-order MASs.
The consensus problem for second-order agents proves
harder. As expected, the conditions sufficient for the con-
sensus of first-order MASs are barely adequate to insure
second-order MASs to achieve consensus [5]. There has
since ensued a multitude of works on MASs of second-
order agents, of which double-integrators provide a notable
case of study. In [6], consensus conditions were given for
the double-integrator MASs with a constant communication
delay over fixed network topology. With given consensus
protocols, necessary and sufficient consensus conditions were
given in [9-11]. Furthermore, robust consensuability prob-
lems were studied in [8, 10, 13], resulting in conditions
that insure double-integrator MASs to maintain consensus
despite that the delay may vary within a range. In this
paper we focus on the robust consensus problem of more
general second-order MASs under an undirected network,
in which the agents are assumed to have real, strictly
unstable poles. We introduce the notion of delay margin for
the consensus problem, and derive explicit bounds on the
delay margin. The results consequently provide conditions
for robust consensuability of such second-order MASs. The
bounds, obtained by optimizing the consensus protocol, show
that the robust consensuability conditions depend critically
on the pole locations of the agents, and on the eigen-ratio
of the network graph, whereas the latter is known to be a
measure of network connectivity.
II. PROBLEM FORMULATION
A. Algebraic Graph Theory Basics
Some basic concepts on algebraic graph theory are intro-
duced. A MAS (or network) is assumed to have N agents
and each agent is assumed to be a node. The communication
topology between agents is denoted by a weighted directed
graph G = (V, E, A) with the set of agents V = {1, ..., N},
set of edges E ⊂ V × V, and a weighted adjacency matrix
A = [a
ij
] with nonnegative adjacency elements a
ij
. The
node indexes belong to a finite index set Γ = 1, 2, ..., N. The
neighboring set N
i
of node i is denoted by N
i
, {j|(j, i) ∈
E}. The in-degree of node i is represented by d
i
=
P
N
i
j=1
a
ij
and the in-degree matrix D , diag{d
1
, d
2
, ...d
N
}. The
Laplacian matrix L of the graph G is defined by L = D−A.
978-1-5386-7345-4/18/$31.00 © 2018 IEEE
Proceedings of the 2018 13th World
Congress on Intelligent Control and Automation
July 4-8, 2018, Changsha, China