Citation

Michalopoulos P, Hourdakis J. Proc. Inst. Mech. Eng. Pt. I: J. Syst. Contr. Eng. 2001; 215(4): 345-355.

Copyright

(Copyright © 2001, SAGE Publishing)

DOI

10.1177/095965180121500407

PMID

unavailable

Abstract

As performance requirements for advanced traffic management systems (ATMSs) increase, the need for new sensors leading to the next-generation traffic management systems is manifested. In spite of the wide array of non-intrusive technologies that are available today, most are not new. Even though earlier technologies continue to improve, most are fundamentally unsuitable for traffic detection, and their widespread use does not appear to be imminent. As a result, most traffic management centres (TMCs) being built today continue to rely on an old but established in-pavement technology, namely loop detectors. Perceived cost and reliability notwithstanding, the major physical disadvantage of loops, as well as of most technologies (intrusive or not), is their inability to detect traffic over a wide area and extract critical traffic measurements such as queues, density, stops and others. In this paper, non-intrusive technologies are briefly reviewed, and particular emphasis is given to the most promising and increasingly accepted in the field, namely machine vision. Since the technology is only recently accepted, representative major field deployment projects are also included.

Language: en