This paper suggests using AIS (Artificial Immune Systems) for automatically detecting and analyzing traffic anomalies due to unpredictable road situations. Various authors [DeCaTi2002], [ForHofSom1996], [TarSkorSoko2003] suggested models and methodologies based on the paradigm of the biological immune system in the framework of network intrusion detection [MykHebLev1994]. We suggest applying an AIS-based recognition model for detecting and controlling pathological road situations, such as traffic jams due to unexpected events. Our approach is based on the idea of defining pathological traffic patterns as antigens, and applying the singular value decomposition (SVD) of matrices of previous abnormal traffic logs to compute matching antibodies. Such antibodies will subsequently be applied to recognize actual traffic patterns as either self (normal traffic), or non-self (pathological traffic). This will be done by (a) using antibodies to map past pathological traffic logs into a subset of a k-dimensional shape space, and cluster them into recognition balls, (b) mapping every new incoming traffic log into a point P of the same shape space, and (c) determine the nature of the incoming traffic log by evaluating its minimum distance recognition balls.

Real-time Detection of Pathological Traffic Situations via AIS / A. Pagnoni, A. Visconti - In: Entwurf komplexer Automatisierungssysteme - EKA 2006 : Beschreibungsmittel, Methoden und Werkzeuge für Entwurf und Zuverlässigkeit von Anwendungen in Automatisierung und Verkehr ; 9. Fachtagung, 29. bis 31. Mai 2006 in Braunschweig / Technische Universität Braunschweig, Institut für Verkehrssicherheit und Automatisierungstechnik. E. Schnieder (Hrsg.) / [a cura di] E. Schnieder. - [s.l] : Institut für Verkehrssicherheit und Automatisierungstechnik, 2006. - ISBN 3-9803363-9-5. (( convegno EKA 2006 tenutosi a Braunschweig nel 2006.

Real-time Detection of Pathological Traffic Situations via AIS

A. Pagnoni
Primo
;
A. Visconti
Ultimo
2006

Abstract

This paper suggests using AIS (Artificial Immune Systems) for automatically detecting and analyzing traffic anomalies due to unpredictable road situations. Various authors [DeCaTi2002], [ForHofSom1996], [TarSkorSoko2003] suggested models and methodologies based on the paradigm of the biological immune system in the framework of network intrusion detection [MykHebLev1994]. We suggest applying an AIS-based recognition model for detecting and controlling pathological road situations, such as traffic jams due to unexpected events. Our approach is based on the idea of defining pathological traffic patterns as antigens, and applying the singular value decomposition (SVD) of matrices of previous abnormal traffic logs to compute matching antibodies. Such antibodies will subsequently be applied to recognize actual traffic patterns as either self (normal traffic), or non-self (pathological traffic). This will be done by (a) using antibodies to map past pathological traffic logs into a subset of a k-dimensional shape space, and cluster them into recognition balls, (b) mapping every new incoming traffic log into a point P of the same shape space, and (c) determine the nature of the incoming traffic log by evaluating its minimum distance recognition balls.
Settore INF/01 - Informatica
2006
Book Part (author)
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/23445
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact