Genetic Algorithms: Application to Fault Diagnosis in Distributed Embedded Systems

Genetic Algorithms: Application to Fault Diagnosis in Distributed Embedded Systems

Pabitra Mohan Khilar (NIT Rourkela, India)
Copyright: © 2014 |Pages: 17
DOI: 10.4018/978-1-4666-4940-8.ch012


Genetic Algorithms are important techniques to solve many NP-Complete problems related to distributed computing and its application domains. Genetic algorithm-based fault diagnoses in distributed computing systems have been a feasible methodology to solve diagnosis problems recently. Distributed embedded systems consisting of sensors, actuators, processors/microcontrollers, and interconnection networks are one class of distributed computing systems that have long been used, staring from small-scale home appliances to large-scale satellite systems. Some of their applications are in safety-critical systems where occurrence of faults can result in catastrophic situations for which fault diagnosis in such systems are very important. In this chapter, different types of faults, which are likely to occur in distributed embedded systems and a GA-based methodology to solve these problems along with the performance analysis of fault diagnosis algorithm have been presented. Nevertheless, the diagnosis algorithm presented here is well suitable for general purpose distributed computing systems with appropriate modification over system and fault model. In fact, this book chapter will enable the reader not only to study various aspects of fault diagnosis techniques but will also provide insight to build robust systems to allow for continued normal service despite the occurrence of failures.
Chapter Preview


Genetic Algorithms are important techniques used for searching large solution spaces for different problem domains in computer science and engineering field such as distributed computing systems (Khilar & Mahapatra, 2007, Barborak, Malek & Dahbura, 1993, Malek, M. 1980, Hakimi & Chwa, 1981, Maeng & Malek, 1981, Blough & Brown, 1999, Elhadef & Becher 2000, Elhadef & Becher 2001). The problem domains are usually consists of a number of metatasks based on different heuristics used to solve the fault diagnosis problems using GA fundamental operators. When the numbers of tasks are very large, the mapping of these tasks to underlying resources is an NP-Complete problem due to large solution space. In order to solve the meta task scheduling problem, different tasks are mapped to various nodes in a distributed computing system. Nodes are usually machines and the connection between nodes is the links.

Fault diagnosis is another important problem in distributed computing systems to locate the faulty node. This is required to identify a set of fault free nodes on which normal function of distributed systems can be successfully executed. As the results of applications are vital and some times safety critical, reliable results will assert in providing correct decision. Although, fault diagnosis tasks are observed as overheads at the initial stage of installation, subsequently they are essential due to deterioration of components of distributed systems. The cost overhead of the system can be very well evaluated and tested for the feasibility of the system in real environment. While considering the execution of normal tasks of a distributed system, the fault diagnosis tasks can be included and executed in an overlapped manner in order to successfully accomplish the normal functioning of the system despite the occurrence of fault.

Distributed embedded systems such as fly-by-wire (FBW), drive-by-wire (DBW) and break-by-wire (BBW) systems are becoming smarter by incorporation of higher computing and communication power along with the new generation automotive design. As the distributed embedded systems become smarter, the data traffic inside the in-house networks also increases dynamically. The total data traffic will be huge by adding up multiple video systems such as displays and cameras. Therefore, the new generation distributed embedded systems need to allow the effective and efficient communication among different subsystems. While the computation are implemented either using a set of processors or microcontrollers, the communication is implemented using multiple networks handling communication of data, varying in criticality as well as bandwidth requirement. The distributed embedded systems such as FBW, DBW and BBW use extensive electronics components in various subsystems such as power train, engine, chassis, body, in-house control for navigation system and equipments for connectivity to the external world. Advanced piloting or driving assistance systems (APDAS) are the innovations added to the traditional airplanes, aircrafts, rockets and vehicles. In fact, APDAS was originally considered under telematics due to the use of more and more new features. Moreover, APDAS systems may need to communicate with safety systems wherever some automatic takeover is required.

Complete Chapter List

Search this Book: