Return to search

Lyapunov-based Control Approaches for Networked Single and Multi-agent Systems with Communication Constraints

Networked control systems (NCSs) are feedback control systems with
the feedback control loops closed via network. The origin of the
term NCSs is from industrial systems where the plant and
controller are often connected through networks. The applications
of NCSs cover a wide range of industries, for example, manufactory
automation, domestic robots, aircraft, automobiles and
tele-operations.

The research activities in NCSs are focused on the following three
areas: control of networks, control over networks and multi-agent
systems. Control of networks is mainly concerned with the problem
of how to efficiently utilize the network resource by controlling
and routing the network data flows. Control over networks is
mainly concerned with the design of feedback control strategies of
control systems in which signals are transmitted through
unreliable communication links. Multi-agent systems deal with two
problems: how the topology of the network connections between each
component influences global control goals and how to design local
control law describing the behavior of each individual to achieve
the global control goal of the whole systems. The objective in
this thesis is to deal with control over networks and multi-agent
systems.

The most challenging problem in the control over networks field is
that the unreliable communication channels can degrade system
performance greatly. The main unreliable properties of networks
are delays and packet loss. In order to deal with this problem, a
Lyapunov-based method has been used to design the sampled-data
stabilization control strategy for a networked single system by
choosing proper delay and packet loss dependent Lyapunov
functional candidates. Linear matrix inequality techniques have
been used to find the sufficient and necessary conditions for the
controller design. Furthermore, the consensus formation control
problem of multiple robotic vehicle systems has been investigated.
The consensus-based design scheme has been applied to the
formation control of multiple wheeled mobile-robot group with a
virtual leader. A novel delay-dependent Lyapunov functional
candidate has been constructed to investigate the convergence of
the system states. The proposed control strategy is experimentally
implemented for multiple wheeled mobile robots under
neighbor-to-neighbor information exchange with group communication
delays involved. In conclusion, through the simulation results and
experimental validations, the proposed new Lyapunov-based control
methods can effectively deal with the networked control systems
discussed in this thesis.

Identiferoai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:NSHD.ca#10222/13158
Date25 November 2010
CreatorsSheng, Long
Source SetsLibrary and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada
LanguageEnglish
Detected LanguageEnglish

Page generated in 0.0021 seconds