FAULT DIAGNOSIS IN NCS UNDER COMMUNICATION
CONSTRAINTS: A QUADROTOR HELICOPTER APPLICATION
K. Chabir, M. A. Sid, D. Sauter
Nancy University, CRAN -CNRS UMR 7039 BP239, 54506 Vandoeuvre Cedex , France.
{karim.chabir, Mohamed-Amine.sid, dominique.sauter}@cran.uhp-nancy.fr
ABSTRACT
In this paper a method for fault diagnosis in quadrotor
helicopter is presented. The proposed approach is
composed of two stages. The first stage is the modelling
of the system attitude dynamics taking into account the
induced communication constraints. Then a robust fault
detection and evaluation scheme is proposed using a
post-filter designed under a particular design objective.
This approach is compared with previous results based
on the standard Kalman filter and gives better results for
sensors fault diagnosis.
Keywords: Networked control systems, Diagnosis,
generation residual, evaluation residual, Quadrator
helicopter.
1. INTRODUCTION
Unmanned Aerial Vehicles (UAV) are receiving a great
deal of attention during the last few years due to their
high performance in several applications such as search
and critical missions, surveillance tasks, geographic
studies and various military and security applications.
As an example of UAV systems, the quadrotor
helicopter is relatively a simple, affordable and easy to
fly system and thus it has been widely used to develop,
implement and test-fly methods in control, fault
diagnosis, fault tolerant control as well as multi-agent
based technologies in formation flight. Navigation and
guidance algorithms may be embedded on the onboard
flight microcomputer/microcontroller or with the
interference by a ground wirelesses/wired controller in
others cases. In our setting the quadrotor is controlled
over real time communication network with time-
varying delays and therefore is considered as a
Networked control system (NCS). In general NCS is
composed of a large number of interconnected devices
(system nodes) that exchange data through
communication network. Recent research on NCS has
received considerably attention in the automatic control
community (Zhang, et al., 01; Tipsuwan and Chow, 03;
Huajing et al., 07; Mirkin and Palmor, 05; Hespanha, et
al., 07; Richard, 03). The major focus of the research
activities are on system performance analysis regarding
the technical properties of the network and on the
controller design schemes for NCS.
However, the introduction of communication networks
in the control loops makes the analysis and synthesis of
NCS complex. There are several network-induced
effects that arise when dealing with the NCS, such as
time-delays (Niculescu, 00; Nilsson, et al., 98; Pan, et
al., 06; Schollig, et al., 07; Dritsas, and Tzes, 07; Yi, et
al., 06; Zhang, et al., 05; Behrooz, et al.,08), packet
losses (Xiong, and Lam, 06; Sahebsara, et al., 07; Yu, et
al., 04; Li, et al., 06) and quantization problems
(Goodwin , et al., 04; Montestruque and Antsaklis, 07;
Frank and Ding, 97). Because of the inherent
complexity of such systems, the control issues of NCS
have attracted attention of many researchers, particulary
taking into account network-induced effects. Typical
application of these systems ranges over various fields,
such as automotive, mobile robotics, advanced aircraft.
The fault diagnosis has become an important
subject in modern control theory (Frank and Ding, 97;
Gertler, 98; Isermann, 06; Stoustrup, and Zhou, 08;
Basseville, and Nikiforov, 93). The study of fault
detection (FD) in NCS is a new research topic, which
gained more attention in the past years. For instance, the
results in (Sauter and Boukhobza, 06; Sauter, et al., 07 ;
Llanos, et al., 07; Chabir, et al., 08; Chabir, et al., 09;
Chabir, et al., 10; Al-Salami, et al., 08) are focus on
networked-induced delays. The problem studied in
(Zhang, et al., 04; Wang, et al., 06) is the analysis and
design of FD systems in case of missing measurements.
The fault detectability and isolability in NCS have been
discussed in (Sauter, et al., 09; Chabir, et al., 09). The
fault tolerant structure is studied in (Ding and Zhang,
07 ; Patton, et al., 07; Kambhampati, et al., 06).
Delays are known to degrade drastically the
performances of a control systems, for this reason,
many works aimed at reducucing the effects of induced
network delays on NCS (Tipsuwan and Chow, 03; Yu,
et al., 04; Li, et al., 06; Goodwin , et al., 04). In the
majority of the studies concerning the stabilization of
networked control systems, the delay is considered to be
constant (Schollig, et al., 07) or bounded (Dritsas, and
Tzes, 07), but the dynamics of the delay corresponding
to the characterization of the network is not taken into
account in general. Thus, it is interesting to estimate the
delay, in order to generate an optimal control, as well as
algorithms of faults detection that take into account the
Proceedings of the Int. Conf. on Integrated Modeling and Analysis in Applied Control and Automation, 2012
ISBN 978-88-97999-12-6; Bruzzone, Dauphin-Tanguy, Junco and Merkuryev Eds.
